The Impact of HeartMate 3 Speed on Outcomes in the Momentum 3 Clinical Trial

Purpose Durable continuous flow left ventricular assist devices (LVADs) operate at a fixed speed set by the clinician. Within Momentum 3, a large multicenter clinical trial of a magnetically levitated centrifugal LVAD, we sought to characterize pump speeds utilized, identify clinical variables associated with speed selection, and determine if early speed setting was independently associated with major adverse events. Methods Momentum 3 was a prospective randomized clinical trial comparing the HeartMate II and HeartMate 3 (HM3) in advanced heart failure. Multivariate linear regression analyses were used to identify baseline clinical factors associated with HM3 speed setting at implant and discharge. The negative binomial model was used to determine if pump speed at implant or discharge from the index hospitalization was independently associated with a composite endpoint of death, transplant for pump thrombosis, and the occurrence of hemocompatibility related adverse events: bleeding due to surgical or non-surgical causes, any stroke, suspected or confirmed pump thrombosis or peripheral arterial thromboembolism. Results There were 514 patients implanted with the HM3 that had speed data available for analysis. Median speed at implant was lower than all other time points through 180 days (Figure). Baseline clinical variables independently associated with higher pump speed at implant were increasing body surface area (BSA, p=0.003) and left ventricular end diastolic diameter (LVEDD, p=0.008), and at hospital discharge were BSA (p Conclusion Within the Momentum 3 clinical trial, HM3 speed was significantly lower at implant compared to other time points up to 180 days. Younger age, male gender, larger BSA and LVEDD were associated with higher pump speeds. Early HM3 speed was not independently associated with adverse events.