New Risk Score May Help Identify LVAD Candidates


A new risk score, specifically calculated for use with the HeartMate II left ventricular assist device (LVAD), may help select appropriate candidates for bridge-to-transplantation and destination therapy. The new analysis, published in the Dec. 19 issue of the Journal of the American College of Cardiology, identified five routinely obtained and reproducible data points predictive of 90-day mortality: age, serum albumin, creatinine, international normalized ratio (INR) and implant center LVAD experience.


There is a greater need for updated risk models for LVAD candidates, according the study's authors. Lead author Jennifer Cower, MD, MS, University of Michigan Health Systems, Ann Arbor, and colleagues write that as LVAD technology has evolved from pulsatile devices to contemporary continuous-flow devices, existing risk scores can no longer provide accurate predictive data for patients undergoing bridge and destination therapy. Thus, the HeartMate II Risk Score (HMRS) was based on the 1,122 patients enrolled in the HeartMate II bridge-to-transplantation and destination therapy clinical trials. Patients were randomized into derivation and validation cohorts. Predictors of 90-day mortality were examined in the derivation cohort (n=583) with logistic regression analysis and applied to the validation cohort (n=539).

The investigators reported that older age, hypoalbuminemia, renal dysfunction, coagulopathy and LVAD center experience were predictive of 90-day mortality, and concluded, "the HMRS might be useful for mortality risk stratification in HMII candidates and may serve as an additional tool in the patient selection process."

Additional Resources


In an accompanying editorial, Wayne C. Levy, MD, FACC, University of Washington, Seattle, concurred with the authors' assessment of the current need for validated risk scores, writing that "evaluation of the mortality risk with medical therapy versus LVAD implantation is imperative as LVADs are expanded into the ambulatory non-inotrope-dependent patients who might have a poor quality of life but might not be a high risk of mortality with medical therapy."


Levy said the HMRS mortality markers validate previous heart failure and LVAD risk models, but he questions the model's widespread applicability, noting that it did not identify patients who were too high-risk for LVAD implantation, but added that it may be effective for low-risk patients.


< Back to Listings