Right ventricular dysfunction (RVD) remains one of the major Achilles’ heels of left ventricular assist device (LVAD) therapy. When present before implantation, it is associated with more complications, longer hospital stays, and poorer survival, and may ultimately influence LVAD candidacy. In these often unstable patients, temporary mechanical circulatory support (tMCS) has increasingly been used as a bridge to implantation, with the aim of stabilizing hemodynamics and improving end-organ perfusion before surgery.

However, its use raises an unresolved question. Large registries have consistently shown that patients who undergo LVAD implantation after receiving tMCS have poorer survival, but it remains unclear whether the support itself contributes to this worse prognosis or simply identifies patients who are more critically ill at baseline. This distinction is not merely theoretical: if the excess risk is confined to the perioperative period, denying LVAD therapy solely because tMCS was required could deprive otherwise suitable candidates of a life-prolonging treatment. Clarifying this issue is the rationale behind the study discussed here.

To address this question, the authors conducted a retrospective multicenter cohort study across four tertiary referral centers, including 201 HeartMate 3 recipients implanted between January 2018 and June 2024, all of whom had objectively defined RVD. The definition of RVD was deliberately stringent, requiring simultaneous confirmation by echocardiography (TAPSE < 16 mm and/or qualitative RV dilation or dysfunction) and invasive hemodynamics (central venous pressure/pulmonary capillary wedge pressure [CVP/PCWP] ratio > 0.54 and/or pulmonary artery pulsatility index [PAPi] < 2.0 with a cardiac index < 2.2 L/min/m²). Patients were stratified according to whether they had received tMCS before implantation (101 with support and 100 without), including intra-aortic balloon pump (IABP), Impella® CP or 5.5, and venoarterial extracorporeal membrane oxygenation (VA-ECMO). Early postoperative morbidity—including the need for and duration of right ventricular assist device (RVAD) support, mechanical ventilation, intensive care unit (ICU) stay, and renal failure—was compared, together with survival up to 3 years, with patients undergoing transplantation censored before completing follow-up. The methodological strength of the study lies in conditional landmark survival analyses at 30, 90, and 180 days and in a Cox proportional hazards model incorporating a time-by-tMCS interaction, thereby separating early risk during the first 90 days from subsequent long-term risk, with adjustment for EUROMACS-RHF score and preoperative intubation.

The first major finding was that the two groups were clearly different at baseline: patients who had received tMCS were substantially more critically ill, with higher-risk INTERMACS profiles, a higher EUROMACS-RHF score (5.8 vs 4.8; 𝑝 = .001), and more severe hemodynamic compromise. This greater baseline acuity translated into increased early morbidity after implantation: RVAD support was required more frequently (53% vs 32%; 𝑝 = .003), and patients experienced longer mechanical ventilation (median 7 vs 2 days; 𝑝 = .04) and ICU stays (15 vs 9 days; 𝑝 = .04). Regarding survival, the tMCS group had significantly greater mortality during the early phase, with differences observed at 30 days, 1 year, and 2 years, whereas the difference was no longer statistically significant at 3 years (68% vs 76%; 𝑝 = .14). The key finding emerged from the conditional survival analysis: once patients survived the first 30, 90, or 180 days, survival curves became virtually superimposable between the two groups (𝑝 = .71, .85, and .63, respectively). Consistently, the time-interaction Cox model showed that tMCS was associated with a markedly increased risk of death during the first 90 days but had no independent effect on mortality thereafter.

The authors therefore conclude that, among LVAD recipients with RVD, preoperative tMCS identifies patients at greater early perioperative risk but does not confer an increased long-term mortality risk among conditional survivors. They consequently emphasize the need to distinguish early procedural hazard from long-term prognosis and caution against interpreting tMCS exposure as evidence of LVAD futility.

COMMENTARY:

The main contribution of this study lies less in what it measures than in how it measures it. Registry data have shown for years that patients undergoing LVAD implantation after preoperative tMCS have poorer survival; what makes this analysis particularly informative is the use of conditional survival, which separates two concepts that are often conflated: the risk of dying during the perioperative phase and the risk of dying over the longer term. By showing that, after the first few months have been survived, the prognosis of patients requiring tMCS becomes indistinguishable from that of patients who did not require support, the authors challenge the deeply ingrained assumption that “preoperative tMCS equals poor prognosis.” Instead, they recast it more accurately: tMCS identifies acuity, not futility. Although apparently subtle, this distinction has direct implications for which patients we accept as LVAD candidates and which patients we exclude.

It is nevertheless important to clarify what type of support is being evaluated, because the title of the study itself may lead to some misunderstanding. The tMCS used in these patients was not dedicated right ventricular support but predominantly left-sided or systemic support: IABP, Impella CP or 5.5, and VA-ECMO. The right ventricle is therefore not directly supported; instead, it may benefit indirectly from left ventricular unloading, reduction in PCWP, and the resulting improvement in RV-pulmonary arterial coupling. This distinction is relevant: the study shows that global stabilization with tMCS in patients with concomitant RVD before LVAD implantation does not compromise long-term prognosis, but this should not be interpreted as a recommendation regarding dedicated right ventricular support, for which the optimal indication, timing, and device remain unresolved.

As with all retrospective evidence, these findings should be interpreted cautiously. The study design inherently carries a difficult-to-avoid selection bias: patients who are so critically ill that they never reach LVAD implantation are not captured, meaning that the favorable long-term prognosis among survivors partly reflects those who were ultimately selected for surgery. Additional limitations include the relatively small sample size and limited number of deaths, which reduce the precision of the estimates and constrain the number of variables that can be incorporated into the analysis, leaving residual confounding from unmeasured factors possible. Moreover, the three support modalities were grouped under a single category despite representing markedly different clinical and physiological profiles. None of these limitations invalidates the overall message, but they do suggest that the findings should be regarded as a robust and well-supported hypothesis rather than definitive evidence.

How should these findings change our practice? Above all, they should influence our approach to critically ill patients with RVD. The study supports the concept that the need for preoperative tMCS, by itself, should neither exclude an otherwise suitable LVAD candidate with RVD nor justify delaying referral. Instead, it should prompt close perioperative surveillance and proactive planning for additional support when this is anticipated to be necessary. Important questions remain unanswered: which patients should receive support, which device should be selected and at what stage, which predictive markers beyond the conventional ones should guide decision-making, and how does right ventricular myocardial deformation influence RV function? These questions will ultimately require prospective studies. In the meantime, this study encourages us to view preoperative tMCS differently: not as the marker of a battle already lost, but as a bridge that may allow a critically ill patient to reach a potentially life-prolonging therapy under the best achievable conditions.

REFERENCE:

Crespo-Diaz R, Hryniewicz K, Samara M, Eckman P, Sun B, Rosenbaum AN, et al. Temporary mechanical circulatory support in left ventricular assist device candidates with right ventricular dysfunction: Acuity without long-term futility. J Heart Lung Transplant. 2026. doi: 10.1016/j.healun.2026.02.1683.