To the Editor:
My coauthors and I read with interest the letter by Lim and Freemantle1 regarding our work calling for careful consideration when attempting to compare nonrandomized trials.2 We appreciate the opportunity to further elucidate some background and clarify any misconceptions.
The selection of the CAP cohort from the BTT + CAP program and the HVAD patients enrolled in the ENDURANCE Supplemental trials to prepare the HVAD Hybrid population was certainly not an attempt to exclude trials with less favorable outcomes as stated by Lim and Freemantle1. Rather, these two datasets represent the most contemporary data available in order to minimize any potential contemporaneous left ventricular assist device (LVAD) patient selection and management differences. Also, the HVAD pump underwent a design change in mid-2011 where sintering was added to the inflow cannula.3 Only patients in the CAP cohort included the current sintered pump design. The patients enrolled in the pivotal BTT cohort were all implanted with pumps with smooth inflow cannulae, as were one-third of the patients enrolled in the ENDURANCE trial.4 The smooth inflow pump design was never commercially available in the US. Moreover, current guidelines recommend deliberate blood pressure management in patients receiving an LVAD, and HVAD patients in the ENDURANCE Supplemental trial were all managed according to current LVAD patient-management practices.
Second, the goal of this brief communication was to highlight the difficulties and pitfalls that can be encountered when attempting to compare nonrandomized trials, and neurologic events were used as an example to highlight the frequently biased and discordant conclusions that stem from failing to take into account study designs, populations, and adverse event definitions. The purpose was not to do an analysis to compare the HVAD to the HM3. Again, as we expressively noted in our article, only a randomized head-to-head trial can appropriately and fairly compare two different devices. The 82.8% survival in the HM3 arm of the MOMENTUM 3 trial was not statistically different from the randomized control in that study at 76.2% (p = 0.16)5; therefore, care should be undertaken in comparisons to the 84% 1 year survival. Despite the fact that data from the BTT + CAP trial demonstrated 79.4% of patients who were alive on device or transplanted at 2 years,6 there remain pitfalls to direct comparisons of survival data. There is a temporal bias where patients implanted in the last 2 to 3 years have benefited from improvements in patient management that were not available when patients were implanted in the BTT + CAP program 6 to 10 years ago.
Finally, we were very transparent and detailed in our discussion of the limitations of the analysis presented, and further reiterated the need for randomized head-to-head trials. We agree that propensity score matching is best used for direct comparisons; however, devoid of patient-level data from the MOMENTUM Trial being made available to us by the sponsor, we could not appropriately match patients that would have been BTT or DT, with the differences among those unique cohorts, to the BTT and DT data available from the HVAD clinical trial programs. Once again, the primary aim of our article was to highlight the difficulties in comparing nonrandomized trials, not to compare the HVAD to the HM3.
We thank Lim and Freemantle1 for their thoughtful review of our article, and for the opportunity to further illustrate the difficulties of attempting to compare nonrandomized trials.
University of Washington
1. Lim HS, Freemantle N. Interpreting comparisons between clinical trials. ASAIO doi: 10.1097/MAT.0000000000000978.
2. 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 2018. doi: 10.1097/MAT.0000000000000931.
3. Slaughter MS, Pagani FD, McGee EC, et al. Use of the HeartWare® ventricular assist system for bridge to transplant: Combined results of the ADVANCE and CAP trial. J Heart Lung Transplant 2013.32: 675–683.
4. 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.
5. 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.
6. Aaronson KD, Silvestry SC, Maltais S, et al. Patients awaiting heart transplantation on HVAD support for greater than 2 years. ASAIO J 2016.62: 384–389.