The ASIS-TC study was a retrospective registry conducted in 16 Spanish HTx centers.4,12 This registry included consecutive adult patients listed for emergency HTx while being supported with intra-aortic balloon pumps, extracorporeal membrane oxygenation, percutaneous VADs or surgically implanted, nondischargeable extracorporeal VADs from January 1, 2010 to December 31, 2020. We excluded patients listed for a second HTx or multi-organ transplantation. The research project was approved by the Ethics Committee for Clinical Research of the Autonomous Community of Galicia (Spain). Informed consent was waived given the retrospective study design.

For the present analysis, we selected from the ASIS-TC study database all patients who were listed for emergency HTx while being treated with an extracorporeal circulatory support system that used a CentriMag pump and was configured for providing LVS or BVS We excluded patients managed with isolated right ventricular support or venoarterial ECMO without LVS.

According to the Spanish organ allocation protocol, patients bridged to HTx on CentriMag LVS or BVS were given the maximum level of waiting list priority, known as “status 0”, throughout the study period.4 Status 0 implied priority for getting the first suitable heart donor retrieved within the whole of Spain. Specific criteria for waiting list prioritization in Spain during the study period are detailed in table 1 of the supplementary data.

Patients were followed up for 1 year after HTx or, if they were discharged alive without undergoing HTx, for 1 year after hospital discharge. Vital status at the end of follow-up was known in all patients.

We analyzed the incidence of relevant clinical outcomes of circulatory support, including HTx, death without HTx, hospital discharge without HTx, and transition to other modes of mechanical circulatory support. We also assessed the incidence of adverse clinical events associated with mechanical circulatory support, such as infection, device malfunction, bleeding, thromboembolism, stroke, cardiac reoperation, hemolysis, and renal failure, as well as in-hospital postoperative outcomes following HTx. Specific definitions of the study outcomes are detailed in the supplementary data. Clinical outcomes were allocated by local investigators.

Continuous variables are presented as means± standard deviation or, when markedly asymmetric, as medians [nterquartile range]. Categorical variables are presented as proportions. Statistical comparisons were conducted by means of the chi-square test, the student t test, and the Mann-Whitney U test, as appropriate.

Multivariable logistic regression was used to control for the effect of potential confounders in the association between the type of support (BVS vs LVS) and the risk of death without HTx. In view of the number of events (n=53), we entered a maximum of 5 covariables in the model, selected on the basis of previous literature, clinical knowledge, and their asymmetric distribution between the 2 study groups, age, ischemic heart disease, previous sternotomies, temporary mechanical circulatory support before CentriMag implantation, and Interagency Registry for Mechanically Assisted Circulatory Support status.

One-year posttransplant survival was assessed by means of the Kaplan-Meier method. Group comparisons were performed by means of the log-rank test. Multivariable Cox regression was used to control for the effect of potential confounders on the association between the type of support (BVS vs LVS) and 1-year posttransplant mortality. In view of the number of events (n=68), we entered a maximum of 7 covariables in the model, selected on the basis of previous literature, clinical knowledge, and their asymmetric distribution between study groups, age, ischemic heart disease, previous sternotomies, Interagency Registry for Mechanically Assisted Circulatory Support status, diabetes, chronic obstructive pulmonary disease, and malignancy.

We also conducted an exploratory comparison of pre- and posttransplant outcomes in patients who underwent device implantation through full median sternotomy vs less invasive approaches.

Statistical analyses were performed with SPSS 25. Statistical significance was set as P <.05.

The ASIS-TC registry included 1036 patients on the waiting list for emergency HTx in 16 Spanish institutions from 2010 to 2020. Among them, 377 patients were managed with extracorporeal circulatory support systems that used the CentriMag pump. We excluded 15 patients managed with venoarterial ECMO with no LVS and 4 patients managed with right ventricular support. Therefore, the study population was composed of 358 patients. A flow chart detailing the selection process of the study population is shown in figure 1.

Figure 1.

Selection of the study population: flow chart. HTx, heart transplantation; MCS, mechanical circulatory support; RVAD, right ventricular assist device; VA-ECMO, venoarterial extracorporeal membrane oxygenation.

(0.14MB).

CentriMag was used for LVS in 213 (59.5%) patients and for BVS in 145 (40.5%) patients. The BVS circuit included an oxygenator in 9 patients.

Devices were implanted through full median sternotomy in 277 (77.4%) patients, and less invasively in 81 (22.6%) patients: through thoracotomies in 79 patients and without chest incisions in 2 patients (transeptal approach). A systemic inflow cannula was placed in the left ventricle in 222 (62%) patients and in the left atrium in 136 (38%) patients. Arterial outflow was cannulated centrally, through the thoracic aorta, in 292 (81.6%) patients and peripherally in 66 (18.4%) patients: in the subclavian/axillary artery 60 (16.7%) and in the femoral artery in 6 (1.7%). A distal arterial cannula for anterograde limb perfusion was inserted in 3 patients who underwent peripheral arterial cannulation.

Among 145 patients who required biventricular support, a venous inflow cannula was inserted directly to the right atrium/thoracic cava vein in 136 (93.8%) and was guided to the same site through transfemoral venous access in 9 (6.2%). A pulmonary outflow cannula was inserted directly into the pulmonary artery in 144 (99.3%) patient and was guided to the same anatomical location through transjugular venous access in 1 (0.7%).

Table 1 shows the baseline clinical characteristics of study patients before device implantation.

Patients managed with LVS were older and had a higher prevalence of ischemic heart disease and cardiogenic shock related to myocardial infarction than patients managed with BVS. Meanwhile, patients who required BVS had a higher prevalence of postcardiotomy shock, previous sternotomy, malignancy, and Interagency Registry for Mechanically Assisted Circulatory Support status 1.

Patients managed with BVS had higher central venous pressure and lower tricuspid annulus plane systolic excursion than patients managed with LVS, as well as lower serum albumin and hemoglobin.

Less invasive surgical approaches were used in 71 (33.3%) patients who received LVS and in 10 (6.9%) patients managed with BVS (P <.001). Peripheral arterial outflow cannulation was also more frequent in the LVS group (n=55, 25.8%) than in the BVS group (n=11, 7.6%; P <.001).

The median duration of support was 15 [0-118] days, with no significant differences according to the type of support (LVS group=15 days vs BVS group=15 days; P=.926). Sixty-six (18.6%) patients were assisted for> 30 days and 5 of them underwent elective replacement of a functioning pump. Figure 2 shows the distribution of the duration of support in the study population.

Figure 2.

Length of mechanical circulatory support in the study population.

(0.15MB).

During the in-hospital period after device implantation, 303 (84.6%) patients underwent HTx and 53 (14.8%) died without receiving a transplant. CentriMag LVS was upgraded to CentriMag BVS in 4 patients; of them, only 1 received a transplant, while 3 died during support.

CentriMag was explanted before a donor was found in 5 (1.4%) patients, among whom 1 died shortly after device removal and 2 transitioned to other types of mechanical circulatory support: 1 patient to the HVAD syste, (Medtronic, Ireland) and another to Cardiohelp (Maquet, Germany) venoarterial ECMO, but both died during the index hospital admission. Only 2 (0.6%) patients were discharged alive from hospital after CentriMag explantation, and both died during the following year.

The cumulative rate of HTx during the index hospitalization was 83.1% in the LVS group and 86.9% in the BVS group (P=.328), while the cumulative rate of in-hospital death without HTx was 16% in the LVS group and 13.1% in the BVS group (P=.455). No statistically significant difference was observed in the rate of in-hospital death without HTx between patients who underwent device implantation through full median sternotomy vs less invasive approaches (14.1% vs 17.3%; P=.475).

By means of multivariable logistic regression, we estimated an adjusted odds ratio for in-hospital death without HTx in patients with BVS vs LVS of 0.94 (95%CI, 0.50-1.78).

Table 2 shows the cumulative incidence of major adverse clinical events associated with mechanical circulatory support. Patients with BVS had significantly higher rates of bleeding (LVS: 31% vs BVS: 44.1%; P=.011), need for blood transfusions (LVS: 67.4% vs BVS: 84.1%; P <.001), hemolysis (LVS: 1.9% vs BVS: 7.6%; P=.008) and renal failure requiring dialysis (LVS: 16% vs BVS: 29%; P=.003) than patients managed with LVS. The incidence of cerebrovascular events was significantly higher in the LVS group (LVS: 14.6% vs BVS: 7.6%; P=.044).

Patients who underwent less invasive device implantation had lower rates of bleeding events (full sternotomy: 39.7% vs less invasive: 24.7%; P=.013), need for transfusions (full sternotomy: 83.4% vs less invasive: 43.2%; P <.001) and cardiac reoperation (full sternotomy: 26.4% vs less invasive: 11.1%; P=.004) than patients who underwent full median sternotomy (table 2 of the supplementary data).

Eighteen (5%) patients experienced device malfunction. Three cases were due to electrical/mechanical failure of the pump, while the remaining 15 were related to the cannulae: 9 cases of thrombosis, 3 cases of displacement and 3 cases of insufficient blood drainage.

No statistically significant differences were observed in the incidence of device malfunction between patients managed with BVS or LVS (2.8% vs 6.6%; P=.105). However, patients who underwent less invasive implantation had a higher cumulative rate of device malfunction than patients implanted through full median sternotomy (9.9% vs 3.6%; P=.023) figure 3.

Figure 3.

Clinical outcomes of temporary mechanical circulatory support with the CentriMag device in the study population.

(0.13MB).

Overall, 8 (44.4%) out of the 18 patients who experienced device malfunction died without receiving a transplant. Specific details of episodes of device malfunction are shown in table 3 of the supplementary data.

Overall, 303 patients underwent emergency HTx during the index hospital admission. Upon HTx surgery, 176 (58.1%) patients were supported with CentriMag LVS and 127 (41.9%) with CentriMag BVS. The mean±standard deviation age of donors was 43.5±12.3 years and 57 (18.8%) were women. Mean ischemic time was 209.4±61.2minutes. No statistically significant differences between BVS and LVS patients were observed regarding donor characteristics (table 1).

During the first posttransplant year, 68 (22.4%) patients died, 36 (20.5%) in the LVS group and 32 (25.2%) in the BVS group. One-year posttransplant survival curves are represented in figure 4.

Figure 4.

Central illustration. One-year posttransplant survival according to the mode of support in place at the time of transplantation: Kaplan-Meier analysis.

(0.15MB).

No statistically significant differences in 1-year posttransplant survival were observed between candidates bridged with LVS or BVS, according to the log-rank test (P=.336). There were no statistically significant differences in 1-year posttransplant survival between patients who underwent device implantation through median sternotomy vs less invasive surgical approaches (78.8% vs 73.1%; P=.325) or between patients supported with the device for> 30 days vs ≤30 days (75.4% vs 78%; P=.670).

By means multivariable Cox́s regression, we estimated an adjusted hazard ratio for 1-year posttransplant mortality in patients with BVS vs LVS of 1.16 (95%CI, 0.69-1.94).

Table 3 shows the incidence of major adverse clinical events during the in-hospital postoperative period after HTx. Patients bridged with BVS showed a statistically significant higher incidence of postoperative renal failure than patients bridged with LVS (34.6% vs 21%; P=.008). No significant differences between the 2 groups were observed regarding the incidence of other in-hospital posttransplant outcomes.

There were no statistically significant differences in the incidence of adverse postoperative outcomes following HTx between patients who underwent conventional vs less invasive device implantation (table 4 of the supplementary data).

The present investigation has some limitations. First, our analysis was based on a general registry of emergency HTx patients, which was not intentionally designed for the specific purpose of the study. Second, our results may be influenced by the selection, observation, information, and confusion biases inherent to observational studies. Third, the registry was not externally monitored, and the allocation of adverse clinical outcomes was left to the discretion of local investigators. Fourth, we analyzed a heterogeneous population of HTx candidates treated in several different Spanish institutions, with variable local protocols and clinical expertise, and over a long period of inclusion and therefore, outcomes could be affected by significant heterogenicity. Of note, we focused specifically on a bridge-to-transplant setting and all patients included in this study were waitlisted with high priority while on temporary mechanical circulatory support. Finally, given the peculiarities of our national organ sharing network, which is characterized by very short waiting times for prioritized HTx candidates, we cannot be sure that our results are reproducible in other circumstances.