Predicting Survival After HeartMate 3 Left Ventricular Assist Device Implantation—Progress Continues

Abstract

As per the National Health and Nutrition Examination Survey, there were nearly 7 million Americans (> 20 years of age) with heart failure (HF) in 2020, with the prevalence predicted to increase 46% from 2012 to 2030.1 Population-based cohorts estimate that 0.2% of HF cases were classified as advanced heart failure (Stage D; advanced structural heart disease and refractory symptoms of HF requiring specialized interventions2). Treatment options for advanced HF include chronic inotropic support, heart transplantation, and mechanical circulatory support. Organ scarcity limits heart transplantation; approximately 3000 individuals are transplanted annually in the United States, with an additional 3,5000 individuals on the transplant waiting list. More than 3,200 individuals have a left ventricular assist device (LVAD) implanted annually, up from 1600 LVADs implanted in 2010.3 The only LVAD currently available in the US is the HM3, indicated for bridge to therapy (BTT; ∼50% of LVAD implantations) and destination therapy (DT). Heart transplantation indeed continues to be the gold standard and definitive treatment for advanced heart failure with 90% one-year survival and median survival of over 12 years. For patients with end stage heart failure on chronic inotropic support, one-year survival is approximately 40%, the picture improves to 85% event-free survival with LVAD implantation.4 There have been several indices used for prognostication and risk-stratification with advanced heart failure: the Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC score), the Seattle Heart Failure Model, and the HeartMate II Risk Score (HMRS).5 These risk models were designed for patients without LVADs (e.g., MAGGIC score) or older LVADs (i.e., HeartMate II [HMII]). Recently, a patient-specific risk score has been developed to predict survival following HeartMate 3 LVAD (HM3) implantation.6 The HM3 Survival Risk Score (HM3RS) was conceived to predict 1- and 2-year survival following HM3 implantation. The risk score includes 6 variables: age, prior cardiac surgery, left ventricular size, right atrial pressure to pulmonary capillary wedge pressure, and two baseline laboratory values (blood urea nitrogen and serum sodium). The score was developed using data from 2200 patients enrolled (n = 1540 patients for derivation cohort; 660 patients for the validation cohort) in the Multicenter Study of MagLev Technology in Patients Undergoing Mechanical Circulatory Support Therapy with HeartMate 3 (Momentum 3) trial. The area under the curve for the receiver-operating characteristic (ROC) analysis was 0.76 at one year and 0.71 at two years, indicating acceptable discrimination of the score in predicting survival. Additional analyses confirmed the HM3RS was able to discriminate between tertiles (i.e., higher-than-average survival, average survival, and lower-than-average survival) of survival following LVAD implantation. There are few overlapping variables between HM3RS and existing risk scores. The Seattle Heart Failure Model includes 20 variables, there are two overlapping variables with the HM3RS: age and baseline sodium. The MAGGIC Score includes 13 variables, with one overlapping variable with the HM3RS: age. The HM3RS was compared with the MAGGIC Score to estimate the difference in survival; the authors noted this can be used to guide patient discussions regarding LVAD implantation verus continued medical therapy. The survival benefit (i.e., reduction in mortality) was 39 to 74% higher with the HM3RS compared to mortality estimated from the MAGGIC score.6 The HMRS, designed for HMII includes 5 variables (age, albumin, creatinine, INR, center volume) with a single overlapping variable: age. These risk scores are essential to provide information to patients and their families, allowing for shared decision making. The simplicity of the HM3RS is ideal with just 6 commonly measured variables. It will be important to assess the accuracy of this new risk score across different patient cohorts.7 Ensuring validation with additional patient cohorts will be essential; the Momentum 3 trial had specific inclusion (i.e., consenting to trial) and exclusion criteria (e.g., history of organ transplant, presence of several end-organ dysfunction or failure) that may reduce generalizability to all patients with advanced HF that are eligible for MCS.8 The authors noted that over 40% of patients enrolled in the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) registry met at least 1 exclusion criterion for the Momentum 3 trial.6 Female as well as racial and ethnic minority and low-income (e.g., patients with Medicaid) patients are traditionally underrepresented in clinical trials; these populations are also less likely to receive advanced therapies for HF including LVAD implantation and heart transplantation. Consequently, the HM3RS may not be valid for these populations although the authors took special efforts to conduct subgroup analyses by sex and race, however, ethnicity nor psychosocial or socioeconomic factors were considered which are associated with outcomes.9-11 Future studies should determine whether HM3RS is generalizable across these patient populations that are underrepresented in clinical trials. The HM3 continues to be the only device implanted at the presented time and the HM3RS advances the field by providing patients and their treating physicians more data points in the decision-making process for this complex patient population.12-16