Combined heart and liver transplantation (CHLT) is still an evolving procedure, representing only 0.3% of all cardiac transplantation in a Unified Network for Organ Sharing (UNOS) database study spanning 28 years. The most common indications are cirrhosis secondary to cardiac disease, usually restrictive cardiomyopathy (most commonly from amyloidosis), and either palliated univentricular congenital heart disease or cardiomyopathy. As such, most patients are in the young adult range (median 28 years), and CHLT is rarely required in childhood. In the pediatric age group, CHLT has been performed in a small number of children with univentricular heart disease and metabolic disorders such as familial hypercholesterolemia.
In the setting of univentricular congenital heart disease palliated with a Fontan circulation, a combination of long-standing elevated right-sided pressures together with chronic hypoxia is thought to lead to liver cirrhosis, portal hypertension, and/or hepatocellular cancer. Hepatocellular cancer is being increasingly reported in this situation and may itself become an indication for liver transplantation (LT) ( Table 12.1 ).
|Cirrhosis||Liver ultrasound||Nodularity, echotexture; difficult to interpret in the setting of venous congestion|
|Transient elastography||Usually artificially elevated with venous congestion but may be useful for monitoring serially|
|Transjugular liver biopsy||Useful in conjunction with MELD XI.|
|Hepatocellular failure||Blood tests||Note: bilirubin elevation useful |
INR prolonged by anticoagulants
Albumin may be low from PLE
Raised ammonia: may be as a result of portosystemic shunts unrelated to cirrhosis.
MELD XI > 20 correlates with outcome
|Portal hypertension||Gastroscopy||Varices, portal gastropathy|
|Liver ultrasound||Reversal of portal flow, splenomegaly, varices (less common)|
|Portal pressure measurements||Direct puncture or via spleen|
|Hepatocellular cancer||Triple-phase CT scan||Use LI-RAD system; note confusion with other nodules; hence may require AFP or other measures|
Generally, the indication for heart transplantation is clear. The decision regarding whether to add LT is less clear because many patients have abnormal liver function tests pretransplant, and the majority of these will improve after heart transplantation. The International Society for Heart and Lung Transplantation has published updated criteria on listing for heart transplantation; however, considerations regarding liver disease are not specifically included. There are, however, some large single-center studies examining the effect of liver disease on outcome after heart transplantation, and these may guide the decision to consider LT. Both model for end-stage liver disease (MELD) and, in particular, modified MELD (international normalized ratio [INR] replaced with albumin, MELD-XI) scores over 20 were significantly associated with increased postoperative complications and mortality after cardiac transplantation. A “liver risk score” based on fibrosis on liver biopsy and MELD-XI was developed and used to compare 12-month mortality after cardiac transplant between those with scores above or below 45. There was a significant difference between uni- and multivariate analyses, suggesting that irreversible liver disease may be predictable at the time of evaluation for heart transplantation, prompting consideration of LT. Of course, there are unlikely to be equivalent data available in the pediatric age group, and hence, adult criteria must be used to guide the decision.
Apart from the degree of hepatic insufficiency, the presence of liver cancer or a metabolic condition (such as amyloidosis or hypercholesterolemia) as the underlying etiology of the heart disease will also prompt requirement for CHLT. In the setting of cirrhosis, careful workup for hepatocellular cancer needs to be undertaken. This may be problematic with the inability to perform magnetic resonance imaging because of the presence of a pacemaker and the risk of diagnostic liver biopsy in the setting of anticoagulation used in many patients with cardiac disease. Triple-phase computed tomography (CT) is therefore used to look for the classical arterial phase enhancement and portal venous washout phase features typical of hepatocellular carcinoma (HCC). There are more stringent imaging criteria for smaller (1–2 cm) compared with larger lesions (2–5 cm). Lesions are classified according to the liver imaging reporting and data system (LI-RADS) system on a scale from LR1 (definitely benign) to LR5 (definitely malignant). This system was developed in 2011 but most recently updated in 2017. Unfortunately, liver nodules are common in post-Fontan patients, and benign nodules can be confused for HCC. In this situation, ancillary features such as raised alpha-fetoprotein and higher central venous pressures may be used.
When the need for an LT is under question, a careful workup including liver biopsy, hepatic venous outflow pressure, pressure measurements to assess portal hypertension, and combined review and consensus between cardiac and LT teams is important.
Timing of Transplantation
CHLT requires careful consideration and extensive workup, with very few centers having performed more than a handful of cases. The supply of available organs for CHLT is limited, which has implications for both patient selection and timing of transplant. In addition, the priority given to CHLT over single-organ recipients varies in different regions and will impact waitlist mortality. In the United States, waitlist mortality of 27% preempted more favorable allocation policies for patients with cardiac disease and high MELD scores in the range of 20 to 29.
Timing of listing is generally based on cardiac criteria, because liver failure and HCC are relatively rare as the indications for CHLT.
Surgical Considerations and Graft Selection
Several factors related to the donor, including age, size matching, degree of illness of the recipient using intensive care unit (ICU) scores, and even donor biomarkers, have been used to try to identify donor-specific factors that may correlate with recipient outcome. Although donor selection criteria overall have been relaxed for heart-only recipients, there is no established donor risk index in common use, partly because many of the factors above have not been shown in recent studies to affect recipient outcome significantly. In pediatrics, where donor acceptance criteria have been stricter in the past, a study has shown that in a high-volume center, outcomes of recipients using grafts rejected by other centers are not inferior.
Graft selection criteria in CHLT are based on expert opinion. Low levels of preexisting donor-specific antibodies, a “good” graft that is 90% to 160% size matched to that of the recipient, and a liver graft that does not require size reduction are factors that have been identified as important for donor selection.
A variety of techniques exist for CHLT depending on whether cardiopulmonary bypass (CPB) is used for both procedures or whether it is ceased after the cardiac transplant with reversal of anticoagulation, with the LT being performed with or without venovenous bypass. Considerations include balancing the risk of CPB with the risks of exposure to the cardiac graft and subsequent metabolic disturbance (hyperkalemia, fluid overload, and acidosis), which occurs with hepatic reperfusion. En bloc transplantation, where the liver and heart are connected by the inferior vena cava, has also been performed, as has sequential transplantation using different donors in several case reports. Beal and colleagues have published a summary of these techniques from 2004 to 2015. More recently, transplant surgeons from the Mayo Clinic, Rochester, Minnesota, have proposed that LT might be performed before heart transplantation in select cases. The main indication is the highly sensitized recipient, often those with congenital heart disease who have had multiple operations and transfusions. For this group, they select donors that are local, of good quality (can withstand a longer ischemic time), and with close human leukocyte antigen matching. Both cardiac and liver dissection are undertaken first and recipient hepatectomy is undertaken while the organs are in transit. Only four transplants, all successful, have been described using this technique to date (RC Daly, SS Kushwaha, B Geske, unpublished data, 2016).
The challenge is to optimize hepatic venous outflow in the setting of potential impaired right ventricular function. As patients are usually managed in the cardiac ICU, the cardiac and liver transplant team assumes responsibility for day-to-day management. LT management is also as per protocol in most centers, with close liaison between the teams.
In the setting of a combined heart-and-liver transplant, immunosuppression regimens used for the heart are followed with the assumption that LTs require less immunosuppression and that the addition of the liver will allow a reduction in immunosuppression compared with isolated heart transplants. The usual regimens include a combination of prednisolone, tacrolimus, and mycophenolate mofetil (MMF), with the prednisolone being weaned after 12 months or so.
Transplant Surveillance: Monitoring and Management
Standard post-transplant monitoring applies to CHLT for rejection, infection, and functional recovery. For rejection, serial measures of cardiac function, tacrolimus levels, liver biochemistry, INR, and routine endomyocardial biopsies in the first 14 days are undertaken. Liver biopsies are performed when clinically indicated for isolated LTs.
Overall, 1- and 5-year post-CHLT survival (85.9% and 74.3% respectively; n = 92) is comparable with that of both isolated heart transplant and isolated LT. The presence of congenital heart disease, often with several previous surgeries, does not adversely impact outcome. The 1-, 5-, and 10-year survival rates for adult patients with congenital heart disease and CHLT are 86%, 83%, and 83%, respectively. Recent data suggest that the addition of a liver allograft results in superior survival to isolated heart transplant. This may be related to the significantly lower episodes of acute rejection in CHLT compared with isolated heart transplant (7.5% vs. 38%; P = .001). There is insufficient literature to provide meaningful outcome data for this procedure in children.
Combined lung-and-liver transplantation (CLLT) is indicated when there is dysfunction of both of these organs and the patient is not expected to survive a single organ transplant. This qualitative statement is guided by outcomes of CLLT from previous series, and the decision must be made in consultation with both teams after an extensive workup. The main indications for this rarely performed procedure are CF, alpha-1 antitrypsin deficiency, pulmonary hypertension, and other primary pulmonary pathologies with secondary or incidental liver disease. The liver disease is usually manifest as portal hypertension with varices and splenomegaly. Most patients undergoing CLLT are adults (median age: 28–40 years).
Timing of Transplantation
Generally, patients listed for lung transplantation are as per lung organ allocation guidelines in the region. In the United States, the Lung Allocation Score (LAS), developed in 2005, aimed to allocate organs to recipients in terms of survival benefit rather than accrued time on the waiting list. Relative and absolute contraindications for lung transplantation were recently updated in a consensus document by the International Society for Heart and Lung Transplantation. Apart from contraindications common to many organ transplants, a notable consideration is chronic infection with highly virulent and/or resistant microbes that are poorly controlled pretransplant. This is important particularly in the setting of CF with chronic antibiotic use. Infections with Burkholderia cenocepacia, Burkholderia gladioli , and multidrug-resistant Mycobacterium abscessus need to be investigated and treated in a center with experience. The liver disease in the most common groups of indications is cirrhotic with portal hypertension, rather than hepatocellular failure, and hence, MELD scores are generally moderate rather than the much higher median of those on the isolated LT waiting list.
Surgical Considerations and Graft Selection
Selection of lung donors was based on stringent criteria including age, chest x-ray findings, smoking history, microbiology, and ischemic time. Multiple studies failed to show any significant impact of smoking, chest x-ray findings, ischemic time greater than 6 hours, or microbiology on recipient outcome, and have allowed extension of these criteria. Age of donors should be between 18 and 64 years. There is an interesting benefit to a female recipient of receiving a male donor, whereas the converse is also true (i.e., a poorer outcome when male recipients receive lungs from female donors). In certain regions where therapeutic manipulation of marginal donors is allowed, “recruitment” strategies (e.g., antibiotic therapy, bronchoscopy and bronchial toilet, and alteration of ventilatory settings) allowed successful use of marginal donors. More recently, ex vivo lung perfusion has also enabled the use of marginal donors for transplantation. How this will fit into the allocation of organs has not yet been established. At present, standard organ-based criteria are used for selection of donors for combined lung-and-liver transplantation.
Lung-and-liver transplantation is performed sequentially by standard techniques, with some centers opting for abdominal dissection before both transplant procedures to minimize the requirement for blood products and exposure of the transplanted lung to fluid overload.
Lung transplantation is undertaken first in most centers because of the sensitivity of the lung graft to ischemia. However, blood loss, hemodynamic instability, and the need for transfusions during LT are thought to predispose to primary graft dysfunction of the lung. A recent series reported from Hannover compared transplantation of liver before lung in 8 recipients with lung before liver in 17 recipients. They reported superior outcomes in the liver-first transplant group in terms of 1- and 5-year outcomes (73% and 73% respectively, compared with 65% and 47% respectively), time on mechanical ventilation of 158 + 124 hours versus 375 + 461 hours ( P = .06), and hospital stays (39 + 17 days vs. 68 + 63 days; P < .05).
Postoperative management is largely coordinated by the lung transplant team. Fluid restriction and vasopressor support targeting a mean arterial pressure of greater than 65 mmHg in adults have been suggested. Infection is a particular focus of lung transplantation, with many patients receiving prolonged courses of broad-spectrum antibiotics post-transplant, directed at pretransplant colonizing organisms. Most patients will experience one or more episodes of infection during their transplant hospitalization.
Immunosuppression is given as per the lung transplant protocol with prednisolone and an antibody (basiliximab) as an induction agent, followed by maintenance with MMF and tacrolimus. Tacrolimus levels targeted are similar to many LT protocols, with a trough of 10 to 12 ng/mL for the first 3 months and 5 to 8 ng/mL thereafter.
Transplant Surveillance: Monitoring and Management
Monitoring for rejection in lung transplants is as per the local protocol and usually involves surveillance bronchoscopies weekly for a month, then two weekly for 3 months, followed by monthly for a year. Most rejection episodes occur in the first 6 months, and most are acute cellular rejection. Liver monitoring is as per standard LT protocol, but episodes of liver rejection appear to be relatively uncommon, with no patients experiencing liver rejection in two series of 8 and 12 patients, respectively.
Pasupneti reported the United States experience of CLLT, which comprised 67 patients over 20 years, most of whom were transplanted after the introduction of the LAS in May 2005. Long-term survival was better overall for CLLT recipients than those who received an isolated lung transplant, with unadjusted 3- and 5-year survival rates of 71.8% versus 64.0% and 66.2% versus 51%, respectively.
The main indication for combined liver-and-pancreas transplantation (CLPT) is in patients with end-stage liver disease who also have insulin-dependent diabetes mellitus. In the pediatric setting, this is most commonly associated with CF. Ongoing diabetes after isolated LT places the patient at risk of cardiovascular disease and may negatively impact graft survival and outcome. On the other hand, the addition of a pancreatic graft adds complexity in terms of managing ischemic times and additional surgery, as well as competing organ allocation policies. For these reasons, a consensus should be reached by the pancreas and LT teams in the region. A recent international survey of 50 pediatric LT (PLT) programs showed a significant variation in opinions regarding indications for CLPT in patients with CF. Although most (94%) would consider offering CLPT in the setting of insulin-dependent diabetes, 50% would consider it in the setting of glucose intolerance and 24% would consider CLPT if objective exocrine insufficiency could be demonstrated. Several groups have reported CLPT in patients with insulin-dependent type 2 diabetes secondary to cirrhosis from various causes, including sclerosing cholangitis, nonalcoholic steatohepatitis, and hepatitis B. The long-term utility and ethics from an organ allocation point of view remain to be established for this approach. Finally, there are isolated case reports of CLPT for rare genetic diseases that affect both liver and pancreas (e.g., Wolcott-Rallison syndrome).
Timing of Transplantation
Timing of transplantation for CLPT is usually dictated by the severity of the liver disease and its complications, as diabetes can continue to be managed medically. There was no significant difference in waitlist times for CF liver-pancreas recipients (median: 189 days; interquartile range [IQR]: 95–412 days) versus CF liver-only recipients (median: 120 days; IQR: 40–342 days; P = .2) in a UNOS database report.
Surgical Considerations and Graft Selection
Guidelines for pancreatic graft selection are somewhat stricter than for liver donors in that donors generally need to be more than 25 kg and up to the age of 45 years without known diabetes mellitus, pancreatic trauma, or history of alcoholism. Vascular thrombosis is more likely to occur in older grafts, whereas donors over 100 kg are more likely to have a fatty pancreas, which negatively impacts graft function. In addition, ABO compatibility is an absolute requirement.
The liver and pancreas can be transplanted en bloc with a C-loop of duodenum or can be implanted separately. In the latter case, the pancreas is heterotopically transplanted onto the iliac vessels in the right lower quadrant. En bloc transplantation described by Starzl offers the advantage of larger vascular anastomoses and no biliary anastomosis, which reduces both surgical difficulty and presumably a reduced propensity for critical post-transplant vascular stenosis. In addition, the pancreas is more physiologically placed, with normal portal venous drainage to the transplanted liver. On the other hand, if there are complications, in particular pancreatic graft thrombosis or leak, the liver graft is less impacted if the pancreas has been heterotopically placed ( Fig. 12.1 ).