Introduction

Combined heart–liver transplantation (CHLT) is a complex surgical procedure indicated for patients with a single underlying disease process causing both heart and liver failure or patients with heart and liver pathology of different etiologies. Accumulating experience with CHLT in the USA has demonstrated acceptable survival compared to orthotopic liver transplantation (OLT) alone and orthotopic heart transplantation (OHT) alone [1, 2]. In most cases of CHLT, the heart is transplanted first as the heart is less tolerant of ischemia, and the heparin needed for cardiopulmonary bypass for the heart transplant can be reversed prior to the liver transplant [3]. However, in specific circumstances, the liver may be transplanted first to protect the heart from donor-specific antibodies (DSA) in patients with high DSA titers.

Sequential heart liver transplantation (SHLT) has been performed when heart and liver allografts could not be obtained from the same donor, or the liver allograft was deemed unacceptable for transplant after its recovery [4, 5]. However, data comparing outcomes of heart and liver transplantation (HLT) performed as a sequential versus a combined procedure are limited. Using data from the United States transplant registry, we compared outcomes of patients undergoing sequential versus combined HLT to determine whether SHLT was associated with an increased risk of adverse outcomes relative to CHLT.

Methods

Following Institutional Review Board approval, de-identified data were obtained from the United Network Organ Sharing (UNOS) Registry. Data from transplants performed in 1987–2018 were included in our analysis, with patient follow-up censored in March 2019. Patients were evaluated for inclusion in the study if they received a first-time OHT either (a) in combination with OLT (combined group, CHLT), or (b) after having been listed for OLT, and having subsequently received OLT (sequential group, SHLT). Patients who received OLT after OHT, but were listed for OLT subsequent to receiving OHT, were excluded from the analysis, as it was unclear whether their OLT was planned to follow OHT (Fig. 1). The primary outcome in the analysis was patient survival from the date of OLT. Secondary outcomes included liver allograft survival and heart allograft survival. Due to variable durations between OHT and OLT in SHLT cases, heart allograft survival was analyzed from the date of the OLT. Actuarial 1-, 3-, and 5-year survival rates were reported for each outcome, and Kaplan–Meier curves with log-rank tests were used to compare patient and graft survival between recipients of combined and sequential HLT. To assess priority for OLT among patients age 12 and older transplanted since February 2002, the final the laboratory model for end-stage liver disease score was compared according to type of OHLT (combined vs. sequential) using a Wilcoxon rank-sum test. To assess priority for OHT since 1998, listing with UNOS Status 1A was compared according to the type of HLT (combined vs. sequential) using Fisher’s exact test. Data analysis was performed using Stata 15\IC (College Station, TX: StataCorp, LP). p < 0.05 was considered statistically significant.

Fig. 1
figure 1

Diagram of included patients

Full size image

Results

The initial cohort included a total of 381 first-time OHT recipients who were also listed for OLT prior to receiving OHT. Twenty-seven patients were excluded as they received OLT prior to OHT. Forty-seven patients received OLT prior to OHT and were not simultaneously listed for both heart and liver at the time they received OLT. In total, 301 CHLT recipients (195/106 male/female, mean age 47 ± 15 years) and six SHLT recipients (all male, mean age at LT 54 ± 6 years) remained for our analysis. The SHLT recipients were listed for both OHT and OLT at the time of the OHT (Fig. 1). Among patients receiving SHLT, median time to OLT was 331 days (interquartile range: 149, 634). Recipient characteristics are summarized in Table 1, and donor characteristics are summarized in Tables 2 and 3. Median MELD laboratory scores did not significantly differ between 275 CHLT recipients (15; interquartile range [IQR]: 10, 19) and four SHLT recipients (11; IQR: 10, 18; p = 0.613) transplanted since the MELD score was incorporated into donor liver allocation. The proportion of patients listed as Status 1A on the heart waitlist did not differ between the CHLT (160/289, 55%) and SHLT (1/4, 25%) cohorts (p = 0.330).

Table 1 Characteristics of combined and sequential heart–liver transplant recipients (N = 307)
Full size table
Table 2 Characteristics of heart donors in combined and sequential heart–liver transplant procedures (N = 307)
Full size table
Table 3 Characteristics of liver donors in sequential heart–liver transplant procedures (N = 6)
Full size table

Patient survival after CHLT was 88% at 1 year, 84% at 3 years, and 82% at 5 years, compared to 83%, 67%, and 50% in the SHLT group (Fig. 2; log-rank test p = 0.010). Liver allograft survival at 1, 3, and 5 years was 88%, 83%, and 82%, respectively, in the CHLT group compared to 83%, 67%, and 50%, respectively, in the SHLT group (Fig. 3; log-rank test p = 0.009). After OLT, heart allograft survival at 1, 3, and 5 years was 86%, 79%, and 74% in the CHLT group, respectively, compared to 83%, 67%, and 50% in the SHLT group (Fig. 4; log-rank test p = 0.0037).

Fig. 2
figure 2

Patient survival

Full size image
Fig. 3
figure 3

Liver allograft survival

Full size image
Fig. 4
figure 4

Heart allograft survival

Full size image

Discussion

Although CHLT remains rare compared to individual organ transplantation, expertise in this procedure and other double-organ transplant procedures has rapidly increased in the USA [6, 7]. The current study demonstrates an increasing use of CHLT with 72% (218/301) of CHLT procedures being performed since 2010. As previously reported by European experts, SHLT is considered a less optimal procedure, commonly performed when a donor liver cannot be transplanted at the same time as a donor heart [3, 4]. Our results seem to substantiate these findings. In the USA, the UNOS Registry includes six cases where a patient received an OLT following an OHT, while listed for OLT at the time they received OHT. Despite the small size of this SHLT cohort, patient and allograft survival in this SHLT cohort were consistently lower than the CHLT cohort. Potential disadvantages of SHLT include a delay in identifying a donor and receiving an OLT, with a median time between OHT and OLT exceeding 6 months in the SHLT cohort, as well as a lack of immunoprotection observed from previous studies from the transplantation of multiple organs from the same donor [6,7,8]. In particular, hepatic transplantation induces immunoprotection to subsequent grafts by reduction in DSAs [8, 9]. When compared with OHT, CHLT decreases the risk of T cell-mediated rejection, antibody-mediated rejection, and cardiac allograft vasculopathy [9, 10]. Similar findings have been demonstrated in other combined solid organ transplantations as combined liver–kidney transplantation has also been shown to impart similar immunoprotection, compared with solitary kidney transplantation [9, 11, 12].

The superior outcomes of CHLT is particularly relevant as this group contained what may be considered higher risk patients for OHT, although UNOS status 1A was not different between two groups. Previous studies of OHT have sown that pre-transplant severity of illness is associated with 1-year post-transplant mortality [13]. Forty-five percent of patients in CHLT cohort of our analysis were critically ill and admitted to the intensive care unit, while the majority of the patients in the SHLT cohort were outpatients. Moreover, the CHLT cohort included patients who also received concurrent kidney transplantation, those requiring mechanical ventilation, and patients requiring heart transplantation related to congenital heart defects, which are known to increase the risk of post-transplant mortality at 1 year. Other factors known to increase the risk of mortality, including advanced donor age, prolonged donor heart ischemia, and female donor to male recipient heart transplantation, did not differ between the two groups [13, 14].

Although there were no differences in the final MELD score between CHLT and SHLT recipients, the final laboratory MELD scores at OLT in the latter group (median of 11, IQR: 10, 18) indicated low priority for OLT. No difference was found in the other risk factors for OLT in our analysis which included ischemic time, highest sodium level, gamma-aminotransferase values, and female sex [12, 15]. Although cardiovascular disease as the cause of death in donors is associated with poor recipient survival, this was not prevalent in the SHLT cohort with our analysis. This would support the claim that the worse outcomes associated with SHLT are not due to patient undergoing SHLT being sicker or higher risk. These worse outcomes that occur are likely due to the lack of immunoprotection offered by multiple grafts from a single donor. This study indicates that, when possible, the heart and liver should be transplanted at the same time, with consideration of passing on single organs and waiting for the availability of both organs from a single donor.

The present study has limitations inherent to its retrospective analysis and the limited sample size of the SHLT cohort. Although the analysis included a relatively large CHLT cohort, only six patients could be included in the SHLT cohort. The study included a significant time period from 1987 to 2019, and therefore evolving changes in clinical practice over the recent years may not be truly reflected in these data. The patients in the two groups were not precisely matched based on clinical data and severity of illness. However, the CHLT group contained higher risk patients based on commonly used clinical factors. Limitations are also derived from the current UNOS data. The authors were not able to evaluate the reason for delay in liver transplant nor the details of the perioperative course. Another limitation is the lack of inclusion of immunosuppression therapies in our analysis due to their unavailability. Despite these limitations, these data suggest that CHLT is superior to SHLT in survival rate and graft survival and may help guide clinicians in making important decisions in a very small and unique patient population. Future research in these patients should investigate the role of immunoprotection in combined organ transplantation from the same donor.