Prediction of Survival After Implantation of a Fully Magnetically Levitated Left Ventricular Assist Device

Mandeep R. Mehra, Aditi Nayak, Alanna A. Morris, David E. Lanfear, Hassan Nemeh, Sapna Desai, Aditya Bansal, Cesar Guerrero-Miranda, Shelley Hall, Joseph C. Cleveland, Daniel J. Goldstein, Nir Uriel, Leway Chen, Stephen Bailey, Anelechi Anyanwu, Gerald Heatley, Joyce Chuang, Jerry D. Estep

Research output: Contribution to journalArticlepeer-review

18 Scopus citations


Background: Clinical trials inform on average efficacy, but individualized risk assessments for outcome prediction are important in guiding treatment implementation. Objectives: The authors developed and validated a patient-specific risk score to predict survival at 1 and 2 years after HeartMate 3 (HM3) left ventricular assist device (LVAD) implantation. Methods: The MOMENTUM 3 (Multicenter Study of MagLev Technology in Patients Undergoing Mechanical Circulatory Support Therapy with HeartMate 3) trial includes 2,200 HM3 LVAD patients in the pivotal trial and Continued Access Protocol study (2014-2018). The authors randomly assigned all patients to a derivation cohort (n = 1,540) or validation cohort (n = 660). Univariate mortality predictors were screened for potential model inclusion, stepwise selection was used to build the multivariable Cox proportional hazards regression model, and performance (discrimination and calibration) was evaluated. Results: Age, prior cardiac surgery (coronary artery bypass grafting [CABG] or valve procedure), lower serum sodium, higher blood urea nitrogen (BUN), small left ventricular size, and right atrial pressure-to-pulmonary capillary wedge pressure (RAP/PCWP) ratio >0.6 were significant risk factors for mortality. Receiver-operating characteristic (ROC) analysis in the validation cohort demonstrated an area under the curve (AUC) of 0.76 (95% CI: 0.70-0.81) at 1 year and 0.71 (95% CI: 0.66-0.77) at 2 years. Calibration between predicted and observed survival of the risk quintiles was high, with Pearson correlation coefficients of 0.986 and 0.994 at 1 and 2 years, respectively. Patients were successfully stratified into tertiles with higher-than-average, average, and lower-than-average survival, and observed mortality risk increased by 2-fold from one tertile to the next. Conclusions: A practical, easy-to-use HM3 Survival Risk Score with 6 components was developed to accurately predict 1- and 2-year survival after HM3 LVAD implantation. The survival risk score can be used to provide individual survival estimates to facilitate shared decision making when considering HM3 LVAD therapy.

Original languageEnglish
Pages (from-to)948-959
Number of pages12
JournalJACC: Heart Failure
Issue number12
StatePublished - Dec 2022


  • HeartMate 3
  • LVAD
  • prediction
  • risk score
  • shared decision making
  • survival


Dive into the research topics of 'Prediction of Survival After Implantation of a Fully Magnetically Levitated Left Ventricular Assist Device'. Together they form a unique fingerprint.

Cite this