To the Editor:
Mahr et al.1 compared neurologic outcomes between patients with Heartmate 3 and Heartware left ventricular assist devices (LVADs). The authors discussed some of the limitations of their analysis, but we believe the limitations of this study merits further examination to help clinicians “make evidence-based and informed decisions.”
Firstly, the authors selectively included the Evaluation of the HeartWare ventricular assist device (HVAD) bridge to transplant trial (ADVANCE) continued access protocol and the ENDURANCE SUPPLEMENTAL studies and excluded other ADVANCE2 and ENDURANCE3 randomized trials. The rationale of this selective exclusion of datasets was not explained. The results will be biased by the exclusion of trials with less favorable outcomes.
Secondly, a study of neurologic complications is certainly important in the field of LVAD therapy. However, it is not clear why other clinically significant events were not studied; most notably device exchange and mortality. The 6 month device exchange rates were 4.5% and 0% in the combined ADVANCE studies4 and Multicenter study of MagLev technology in patients undergoing mechanical circulatory support therapy with HeartMate 3 (MOMENTUM 3),5 respectively, despite 21.4% of patients undergoing transplantation (or recovered) in the former, which would have reduce the denominator, as highlighted by the authors. The 1 year survival in the combined ADVANCE studies was 84%,4 compared with 2 year survival of 82.8%, despite a similar proportion of patients undergoing heart transplantation (21.2%) in MOMENTUM 3.6
Thirdly, it is axiomatic that randomized controlled trials are considered the gold standard approach for estimating the effects of treatments on outcomes, as random allocation ensures that treatment status will not be confounded with either measured or unmeasured baseline characteristics. Propensity score matching is commonly used in the absence of randomization. Conditional on the propensity score, the distribution of observed baseline covariates is similar between treated and untreated subjects. Propensity score matching will, on average, result in measured baseline covariates being balanced between treatment groups. Propensity score matching or some other reasonable technique for making comparison outside randomized trials should be used to minimize bias when making comparison between studies.
Finally, we agree with the authors that careful consideration of statistical designs and analyses are essential, and arguably even more pertinent in such indirect cross-trial comparisons that are fraught with biases.
Hoong Sern Lim
University Hospitals Birmingham NHS Foundation Trust
Queen Elizabeth Hospital Birmingham
Birmingham, United Kingdom
University College London
London, United Kingdom
1. Mahr C, Pham DT, Mokadam NA, et al. Interpeting neurologic outcomes in a changing trial design landscape: An analysis of HVAD using a hybrid intention to treat population. ASAIO 2019.65: 293–296.
2. Aaronson KD, Slaughter MS, Miller LW, et al.; HeartWare Ventricular Assist Device (HVAD) Bridge to Transplant ADVANCE Trial Investigators: Use of an intrapericardial, continuous-flow, centrifugal pump in patients awaiting heart transplantation. Circulation 2012.125: 3191–3200.
3. Rogers JG, Pagani FD, Tatooles AJ, et al. Intrapericardial left ventricular assist device for advanced heart failure. N Engl J Med 2017.376: 451–460.
4. Slaughter MS, Pagani FD, McGee EC, et al.; HeartWare Bridge to Transplant ADVANCE Trial Investigators: HeartWare ventricular assist system for bridge to transplant: Combined results of the bridge to transplant and continued access protocol trial. J Heart Lung Transplant 2013.32: 675–683.
5. Mehra MR, Naka Y, Uriel N, et al.; MOMENTUM 3 Investigators: A fully magnetically levitated circulatory pump for advanced heart failure. N Engl J Med 2017.376: 440–450.
6. Mehra MR, Goldstein DJ, Uriel N, et al.; MOMENTUM 3 Investigators: Two-year outcomes with a magnetically levitated cardiac pump in heart failure. N Engl J Med 2018.378: 1386–1395.