Five-Year Outcomes in Patients With Fully Magnetically Levitated vs Axial-Flow Left Ventricular Assist Devices in the MOMENTUM 3 Randomized Trial

Abstract

Importance: Although durable left ventricular assist device (LVAD) therapy has emerged as an important treatment option for patients with advanced heart failure refractory to pharmacological support, outcomes, including survival, beyond 2 years remain poorly characterized.

Objective: To report the composite end point of survival to transplant, recovery, or LVAD support free of debilitating stroke (Modified Rankin Scale score >3) or reoperation to replace the pump 5 years after the implant in participants who received the fully magnetically levitated centrifugal-flow HeartMate 3 or axial-flow HeartMate II LVAD in the MOMENTUM 3 randomized trial and were still receiving LVAD therapy at the 2-year follow-up.

Design, setting, and participants: This observational study was a 5-year follow-up of the MOMENTUM 3 trial, conducted in 69 US centers, that demonstrated superiority of the centrifugal-flow LVAD to the axial-flow pump with respect to survival to transplant, recovery, or LVAD support free of debilitating stroke or reoperation to replace the pump at 2 years. A total of 295 patients were enrolled between June 2019 to April 2021 in the extended-phase study, with 5-year follow-up completed in September 2021.

Exposures: Of 1020 patients in the investigational device exemption per-protocol population, 536 were still receiving LVAD support at 2 years, of whom 289 received the centrifugal-flow pump and 247 received the axial-flow pump.

Main outcomes and measures: There were 10 end points evaluated at 5 years in the per-protocol population, including a composite of survival to transplant, recovery, or LVAD support free of debilitating stroke or reoperation to replace the pump between the centrifugal-flow and axial-flow pump groups and overall survival between the 2 groups.

Results: A total of 477 patients (295 enrolled and 182 provided limited data) of 536 patients still receiving LVAD support at 2 years contributed to the extended-phase analysis (median age, 62 y; 86 [18%] women). The 5-year Kaplan-Meier estimate of survival to transplant, recovery, or LVAD support free of debilitating stroke or reoperation to replace the pump in the centrifugal-flow vs axial-flow group was 54.0% vs 29.7% (hazard ratio, 0.55 [95% CI, 0.45-0.67]; P < .001). Overall Kaplan-Meier survival was 58.4% in the centrifugal-flow group vs 43.7% in the axial-flow group (hazard ratio, 0.72 [95% CI, 0.58-0.89]; P = .003). Serious adverse events of stroke, bleeding, and pump thrombosis were less frequent in the centrifugal-flow pump group.

Conclusions and relevance: In this observational follow-up study of patients from the MOMENTUM 3 randomized trial, per-protocol analyses found that receipt of a fully magnetically levitated centrifugal-flow LVAD vs axial-flow LVAD was associated with a better composite outcome and higher likelihood of overall survival at 5 years. These findings support the use of the fully magnetically levitated LVAD.

Trial registration: ClinicalTrials.gov Identifier: NCT02224755 and NCT03982979.

Conflict of interest statement

Conflict of Interest Disclosures: Dr Mehra reported receiving payment made to his institution from Abbott for consulting during the conduct of the study; personal fees from Baim Institute for clinical research as a clinical events committee member, Natera for academic consulting, Paragonix for academic consulting, Transmedics for serving as an advisory board member, NupulseCV for serving as an advisory board member, FineHeart for serving as an advisory board member, Leviticus for serving as an advisory board member, personal fees from Janssen for serving as a steering committee member, Moderna for serving as a clinical endpoint committee member, and Mesoblast for serving as a data and safety monitoring board member outside the submitted work. Dr Goldstein reported receiving personal fees from Abbott for consulting and personal fees from Abiomed for serving as a consultant and speaker outside the submitted work. Dr Cleveland reported receiving grants from Abbott Medical during the conduct of the study and personal fees from Abbott for serving as a medical chair of the clinical events committee on the ARIES trial and grants from Medtronic outside the submitted work. Dr Cowger reported receiving personal fees from Abbott for serving on a publication committee as a consultant during the conduct of the study and nonfinancial support from Abbott for serving on an advisory board, receiving travel-related compensation, serving as a speaker (all for left ventricular assist devices) and serving on a steering committee for Tendyne and Cephea valves; receiving personal fees from Medtronic for serving as a study national primary investigation and consultant (HVAD device); serving on a data and safety monitoring board for Bivacor; and receiving personal fees from Corwave for serving as a consultant outside the submitted work. Dr Hall reported receiving personal fees for consulting from Abbott, Abiomed, Medtronic, CareDx, and Natera outside the submitted work. Dr Salerno reported receiving personal fees from Abbott for serving as a consultant and speaker during the conduct of the study and personal fees from Abbott outside the submitted work. Dr Naka reported receiving speaker fees from Nipro outside the submitted work and serving as a consultant for the past 36 months without payment for Abbott, Zimmer-Biomet, and CryoLife. Dr Horstmanshof reported receiving personal fees from Abbott Medical outside the submitted work. Dr Chuang reported receiving salary as a paid employee of Abbott during the conduct of the study. Dr Wang reported receiving salary for being an employee of Abbott during the conduct of the study. Dr Uriel reported receiving grants from Abbott, Abiomed, and Fire 1 and serving on a medical advisory board Livemetric, Leviticus, and Revamp outside the submitted work.

Figures

Figure 1.. Flow of Patients in a Trial of 5-Year Outcomes in Patients With Fully Magnetically Levitated vs Axial-Flow Left Ventricular Assist Devices

The investigational device exemption trial followed up patients for 2 years and the postapproval extended-phase study followed up patients still receiving left ventricular assist device (LVAD) therapy up to 5 years. aSpecific inclusion and exclusion criteria were not documented at screening. bPatients were randomized in a 1:1 ratio and randomization was stratified by study center and blocked to maintain the 1:1 ratio over time. cIncarceration, nonadherence, suicide attempt/depression, or cancer diagnosis. dA total of 178 patients with magnetically levitated centrifugal-flow pumps and 117 with axial-flow pumps were consented then enrolled in the extended-phase study protocol. Limited 5-year outcome data were obtained in 63 patients in the centrifugal-flow pump group and 87 in the axial-flow pump group who were not successfully enrolled in the extended-phase study. There were 17 patients in the centrifugal-flow pump group and 15 in the axial-flow pump group who did not provide consent for extended follow-up and were considered to be alive on the date of refusal and then censored in the analysis.

Figure 2.. Composite End Point and Overall Survival in a Study of 5-Year Outcomes in Patients With Fully Magnetically Levitated vs Axial-Flow Left Ventricular Assist Devices (LVADs)

A, Comparison of survival to transplant, recovery, or LVAD support free of debilitating stroke (Modified Rankin Scale score >3) or reoperation to replace the pump 5 years after the implant. The 5-year event-free survival was 54.0% (95% CI, 48.5%-59.1%) in the magnetically levitated centrifugal-flow pump group and 29.7% (95% CI, 24.7%-34.9%) in the axial-flow pump group. Patients receiving transplants for reason other than device malfunction are censored at time of transplant assuming there was no prior pump replacement or debilitating stroke. B, Five-year survival was 58.4% (95% CI, 52.8%-63.6%) in the centrifugal-flow pump group and 43.7% (95% CI, 37.6%-49.6%) in the axial-flow pump group. Patients who received transplants were censored at time of transplant. The median (IQR) LVAD support duration was 2.01 (0.94-5.0) years for the centrifugal-flow pump group and 1.86 (0.75-3.98) years for the axial-flow pump group. In the extended phase after 2 years, a total of 477 patients contributed data to these analyses.

Figure 3.. Serious Adverse Events in a Study of 5-Year Outcomes in Patients With Fully Magnetically Levitated vs Axial-Flow Left Ventricular Assist Devices

Serious adverse events are defined as those causing death or congenital abnormality or birth defect or a life-threatening illness or injury that results in permanent disability, requires hospitalization, or prolongs a hospitalization and/or requires intervention to prevent permanent injury or damage. Serious adverse events occurring after 2 years were collected in 178 patients in the magnetically levitated centrifugal-flow pump group and 117 patients in the axial-flow pump group. Other secondary end points are shown in eTable 6 and 7 in Supplement. aRate ratios and P value from Poisson regression. bIncludes encephalopathy, seizure, transient ischemic attack, and other neurological events other than stroke.

Figure 4.. Cause-Specific Mortality 5 Years After Implant in a Study of Patients With Fully Magnetically Levitated vs Axial-Flow Left Ventricular Assist Devices

Comparison of deaths due to hemocompatibility-related events, heart failure, infection, and other causes. In the extended phase after 2 years, a total of 477 patients contributed data to the analysis. aDifference shown for magnetically levitated centrifugal-flow pump minus axial-flow pump values. bHazard ratios and P values were calculated from the Fine-Gray model. cOther causes are provided in eTable 8 in Supplement 2. The most common other causes included ventricular arrhythmia or cardiac arrest, trauma, hepatic or kidney failure, respiratory failure, and driveline or power cable disconnection.