EASL Clinical Practice Guidelines

Medical complications

Early post-transplant and long-term follow-up

The majority of deaths occur within the early post-liver transplant period. The causes of death and graft loss vary according to the time period from LT. Infections, intra- and perioperative surgical complications account for almost 60% of deaths or graft losses in the first operative year, whereas de novo malignancies and cardiovascular diseases are the major reasons for deaths thereafter.

Recurrence of the underlying liver disease, in particular hepatitis C infection, is a significant growing cause of late allograft dysfunction. The prevalence of acute and chronic rejection has been constantly declining over the previous years, mainly due to new potent immunosuppressive regimens. Approximately 15–30% of LT recipients develop one or more episodes of acute cellular rejection, which can be successfully treated with increased immunosuppression in almost all patients. In contrast, chronic (ductopenic) rejection can be effectively treated only in early cases and may lead to graft loss. However, the rate of graft loss due to ductopenic rejection has significantly decreased to less than 2%. Therefore, acute or chronic rejections are uncommon complications leading to allograft dysfunction or death.

Management of HCV recurrence

Hepatitis C recurrence is universal after LT in patients with detectable HCV RNA [380]. Progression of hepatitis C is accelerated after LT and HCV-infected recipients have a reduced graft and patient survival when compared to HCV negative recipients [381]. Around one third of HCV-infected LT recipients will suffer an aggressive HCV recurrence after LT and are at risk of clinical decompensation and graft loss [[28], [382]]. Follow-up of patients with recurrent hepatitis C is usually performed with protocol liver biopsies, which are used to assess the degree of necroinflammation and the fibrosis stage, as well as to exclude other potential causes of graft damage (rejection, drug toxicity). Early identification of patients with progressive hepatitis C is crucial and liver biopsy, hepatic venous pressure gradient (HVPG) measurement or transient elastography (TE) performed one year after LT have shown an excellent ability to identify “rapid fibrosers” [[383], [384], [385]]. Indeed, the presence of significant fibrosis (F ⩾2 METAVIR), portal hypertension (HVPG ⩾6 mmHg) or high TE values (>8.6 kPa) one year after LT are excellent predictors of graft loss. These patients should be considered for early antiviral therapy. TE can be repeated over time to assess fibrosis progression without the need to use an invasive test.

Recommendation

HCV treatment after LT

When eradication of HCV is not feasible before LT, the graft becomes infected universally and immediately after the procedure. HCV infection after LT is characterized by an accelerated fibrotic progression towards chronic hepatitis and cirrhosis. Fibrosis is the main consequence of an imbalanced repair process occurring in the liver in response to the viral injury.

Antiviral therapy after the graft becomes infected can be initiated at early stages (pre-emptive therapy) or once liver damage has already been established [386]. During the first months following LT, patients are still under strong immunosuppression, at risk of opportunistic infections or surgical complications and undergoing treatment with multiple drugs. Several trials assessing pre-emptive therapies with PegIFN and RBV in early phases after LT reported very poor efficacies and poor tolerability due to the presence of renal impairment, infections and cytopenia. To date, the most common and classical approach to treat hepatitis C after LT has been to start antiviral therapy once histological damage is confirmed [[27], [28]]. Overall SVR rates with PegIFN plus RBV have been shown to be low (30–40%) after transplantation, mainly explained by the high rates of treatment discontinuation (20–38%), dose drug reductions (66–73%) and poor tolerance observed in these patients. Liver transplant recipients are prone to haematological toxicity (particularly anaemia). Although the risk of rejection is not high, it has been reported to occur in ∼5% of IFN-treated patients. Different series have evaluated the safety and efficacy of triple therapy with first generation protease inhibitors (telaprevir or boceprevir) in over 300 HCV-infected liver transplant recipients [[387], [388], [389]]. Most of these patients had already significant fibrosis in the graft (⩾F2) or fibrosing cholestatic hepatitis at time of treatment initiation and around half of them were already treatment-experienced after LT. Overall, reported SVR12 rates ranged between 48% and 59%. Nevertheless, the rate of SAEs leading to treatment discontinuation (13–26%) was high; anaemia was the most frequent adverse event and the use of erythropoietin and the need for RBV dose reduction were almost universal. Only one prospective study has evaluated the safety and efficacy of triple therapy with telaprevir in genotype 1-infected patients with less severe recurrence: final results suggest a good safety profile and improved efficacy, with an SVR12 of 72% (53 of 74 patients) [390]. Since telaprevir and boceprevir are substrates and inhibitors of the CYP3A4 system (as well as P-glycoprotein transporter), patients need significant adjustments of CsA and Tac doses; drug levels need to be monitored closely when treatment is initiated as well as when the protein inhibitors are interrupted [391].

Currently, all HCV-infected liver transplant patients should undergo treatment with IFN-free regimens, if available.

The safety and efficacy of sofosbuvir plus RBV administered for 24 weeks was investigated in a phase II pilot single-arm study in 40 patients (naïve or treatment-experienced) with hepatitis C recurrence at least 6 months after LT [392]. Patients with decompensated cirrhosis were excluded. SVR24 was reached in 70%. Despite the small sample size the safety profile was good and most reported side effects were mild. Similarly, a compassionate use program of sofosbuvir plus RBV in patients with severe hepatitis C recurrence after LT was initiated in 2013. Results from the first 104 patients (including some with fibrosing cholestatic hepatitis) were reported recently [393] and indicated SVR12 rates higher than 50%. More importantly, patients’ clinical condition was considered to improve significantly (reduction or disappearance of clinical decompensation, significant amelioration of liver function) in around 2/3 individuals. Both viral clearance and clinical improvement were significantly higher in individuals with early severe recurrence (diagnosed during the first year after LT) than in those with advanced cirrhosis years after LT. These results can be considered excellent taking into account the poor outcomes of the disease.

The safety and efficacy of ABT450/r, ombitasvir, dasabuvir and RBV was assessed in 34 genotype 1-infected liver transplant recipients. Patients were treatment naïve and had mild fibrosis. Safety was good and SVR12 rates were very high (97%). Due to the interactions of ABT450/r with Tac and CyA, changes in immunosuppression were necessary during antiviral therapy [394].

Preliminary data from an ongoing clinical trial assessing the efficacy and safety of the fixed-dose combination of sofosbuvir and ledipasvir with RBV for 12 or 24 weeks were recently presented. The study included treatment-naïve and treatment-experienced patients with genotype 1 or 4 infection, with all fibrosis stages (F0 to F4) including patients with Child-Pugh B and C decompensated cirrhosis [395]. The SVR rates were 97% (108/111) in F0-F3 patients, 96% (49/51) in Child-Pugh A patients, and 84% (37/44) in Child-Pugh B patients. There were no differences in efficacy between 12 and 24 weeks of therapy and the combination had an excellent safety profile. MELD scores at week 4 post-treatment improved in the majority of Child-Pugh A and B patients who achieved viral clearance.

Data from real-life cohorts with a combination of sofosbuvir and simeprevir with or without RBV for 12 weeks were recently reported. SVR12 was achieved in 91% (60/66) of patients infected with genotype 1, most of whom were treatment-experienced with one third having advanced fibrosis or cirrhosis [396]. In the TARGET real-life cohort study, in which most patients were treatment-experienced and more than half had cirrhosis, the combination of sofosbuvir and simeprevir yielded a 90% (61/68) SVR4 rate [397].

The impact of HCV clearance in the transplant setting is high due to the accelerated course of the disease. The latter is particularly relevant in individuals with advanced liver disease: liver fibrosis can regress, HVPG values improve and at the end patient survival is better compared to non-responders or non-treated individuals [[398], [399]]. Although these data are derived from IFN-based treated cohorts, they are most likely applicable for all treatments, regardless of the type of antiviral regimen used. This is further supported by data from the sofosbuvir compassionate program discussed above.

The development of direct-acting antivirals is the beginning of a new era for treatment of HCV patients.

Recommendations

Prevention and treatment of HBV recurrence

Before the use of the hepatitis B immunoglobulin (HBIG) in the early 1990s, more than 75%–80% of liver grafts became infected in HBV-infected patients. The risk for graft infection was high (∼70%) among individuals with HBV-related cirrhosis, intermediate (∼40%) among those with HDV-related cirrhosis, and low (<20%) among patients with acute liver failure. High levels of HBV DNA at the time of LT is the most important determinant of hepatitis B recurrence [400].

In the last two decades, the availability of HBIG and NUCs have changed the prognosis for patients with HBV infection who underwent LT, by reducing recurrence of infection. Patients undergoing LT for HBV-related cirrhosis have currently excellent long-term outcomes, with 5-year survival rates equal to or greater than 80% [[18], [401]]. These figures are comparable or even superior to those of individuals who received LT for other chronic liver diseases.

Preventing HBV recurrence after LT

Samuel et al. [400] reported a large reduction in graft infection (from 75% to 33%) and an increase in 3-year survival (from 54% to 83%) among patients given long-term therapy with parenteral HBIG, starting at the time of LT. HBIG probably acts through several different mechanisms, such as binding to circulating virions, blocking the HBV receptor in hepatocytes, and promoting lysis of infected cells by antibody-dependent cell-mediated cytotoxicity. However, monotherapy with HBIG still resulted in unacceptable rates of hepatitis B recurrence in individuals with detectable levels of HBV DNA at the time of LT. Thus, the current strategy to prevent recurrence of HBV infection after LT includes a combination of HBIG and NUCs (usually lamivudine), with a success rate higher than 90% [[402], [403], [404]]. Among more than 2162 patients treated with variable HBIG regimens and lamivudine, HBV infection recurred in only 143 patients (6.6%) during a follow-up period of 6–83 months [402]. Moreover, a meta-analysis of six studies found that combining HBIG and lamivudine (compared to only HBIG) reduced HBV recurrence and HBV-related death more than 10-fold [405]. The optimal strategy for patients who have developed lamivudine resistance is not well-established, but tenofovir is used in this situation. In the setting of LT, nephrotoxicity should be always considered and renal function should be carefully monitored because of the concomitant use of CNIs.

Due to the high cost of HBIG, several studies have assessed the efficacy of lower doses of HBIG, intramuscular or subcutaneous injections, or even HBIG withdrawal in selected patients. All these minimized prophylactic strategies, in combination with NUCs, have effectively prevented recurrence. Gane et al. [406] reported a recurrence rate of only 4% 5-years after patients were given intramuscular injections of HBIG (400–800 IU/month) in combination with lamivudine. Importantly, this approach reduced costs by as much as 90%, compared with the high-dose intravenous HBIG regimens. A short course of HBIG plus lamivudine, followed by lamivudine monotherapy, was effective in patients with undetectable levels of HBV DNA at the time of transplantation [407]. Thus, withdrawal of HBIG, with NUCs appears to be a feasible approach for HBeAg-negative patients who undergo LT with undetectable levels of HBV DNA.

As NUCs therapies have become more efficacious, the question whether HBIG is needed at all has been debated. The largest study published recently by Fung et al. [408] using prophylaxis with NUCs (no HBIG) would suggest that this is a feasible strategy: the rate virological relapse in 176 patients treated with entecavir at 3 years was 0%. Preliminary safety and efficacy data with tenofovir and emtricitabine with or without HBIG have also been reported [409]. Some of these patients treated only with NUCs may have reappearance of HBsAg in the absence of detectable HBV DNA or ALT elevation. This opens the problem of deciding if what we want is prevention of graft infection (which would necessitate the use of HBIG) or just to control recurrent infection (in this case HBIG is probably not necessary) [409]. Since specific prophylaxis for HDV reinfection is not available, the most effective strategy to prevent HDV reinfection is the the standard HBV prophylaxis with HBIG and antiviral therapy.

Recommendations

Treatment of HBV recurrence after LT

Recurrence is characterized by reappearance of HBsAg in serum and quantifiable levels of DNA; it is frequently associated with clinical evidence of recurrent disease. The aim of therapy is to control HBV replication over time, to prevent graft loss. Entecavir might be a better choice for individuals with renal failure. Tenofovir is the best alternative for patients with lamivudine resistance [17].

Recommendation

Prophylaxis in patients receiving livers from anti-HBc positive donors

Cholongitas et al. [179] reviewed 38 studies on the use of livers from anti-HBc positive donors in 788 HBsAg negative recipients. The probability of de novo HBV infection of recipients who did not receive immunoprophylaxis was as high as 47.8% in seronegative patients (anti-HBc negative and anti-HBs negative) and 15.2% in patients with serologic markers of past infection (anti-HBs and/or anti-HBc positive); HBV infection was particularly low (1.5%) in anti-HBc and anti-HBs positive recipients. Post-transplant immunoprophylaxis against HBV significantly reduced the probability of de novo infection, from 28% (no prophylaxis) to 8.2% (prophylaxis).

Different post-LT prophylaxis strategies (HBIG only, lamivudine only, a combination of HBIG and lamivudine, and/or HBV vaccination) have been tested in patients who received livers from anti-HBc positive donors. However, lamivudine monotherapy is the best cost-effective treatment due to the low rates of graft infection (<3%). HBIG should not be used in HBsAg negative patients, who received a liver from an anti-HBc positive donor.

Recommendations

Management of patients transplanted for alcoholic liver disease

Post-transplant outcomes for patients undergoing LT for alcoholic liver disease are good, similar to individuals transplanted for other forms of liver disease [410]. The natural history of alcoholism is often a relapsing-remitting pattern of alcohol use, which means that a thorough assessment of the disease before indication of a LT and a follow-up after the procedure are crucial to achieve success. Due to the lack of a generally accepted definition of alcohol relapse the recurrent rates are highly variable ranging between 10–50% [[411], [412]], which is, as expected, significantly lower compared to non-transplanted population. Most of these studies defined relapse as any alcohol use regardless of alcohol amount. It has shown that the majority of patients remain abstinent or consume only small amounts of alcohol following LT [413]. Long-term studies have demonstrated that occasional or moderately heavy drinking does not impact graft function or patient survival. Nearly 10–20% of relapsers will have a harmful drinking pattern [414]. Despite differences in the literature, most studies suggest that harmful drinking after LT is associated with a decreased survival [[411], [415], [416]]. Lower survival in recidivists is very clear in studies with 10 years of follow-up [[42], [415]]; however, in studies with 5 years of follow-up this difference is less evident [[417], [418]]. Therefore, all patients with a positive history of alcoholic liver disease should be encouraged to remain completely abstinent from alcohol post-LT and to enter psychiatric therapy or counselling if they relapse into regular alcohol consumption in the post-operative course.

Since patients with alcoholic liver disease are very frequently heavy smokers, it is important to remember that higher incidence of oropharyngeal neoplasms: a complete examination of the oral tract should be performed before transplantation and also periodically after transplantation.

Recommendations

Recurrence of non-alcoholic fatty liver disease

NAFLD and NASH, either de novo or recurrent, are commonly seen after LT [[419], [420]]. BMI prior and following LT, diabetes mellitus, arterial hypertension and hyperlipidaemia are the major risk factors for post-LT NAFLD/NASH. New onset or recurrent NAFLD/NASH may present with elevated serum transaminases and/or typical features on ultrasound; however, in order to distinguish NAFLD/NASH from other causes of elevated liver tests a liver biopsy may be required.

So far, there is no evidence that recurrent NASH may lead to significant fibrosis or even liver cirrhosis; however, most of these studies are limited by short follow-up periods [421]. No specific recommendations regarding prevention and treatment of recurrent NASH can be made, except to avoid excessive weight gain and to control diabetes and dyslipidaemia.

Although there are no strong data suggesting a specific immunosuppressive strategy for patients undergoing LT for NASH cirrhosis, minimizing corticosteroids seems to be prudent.

Recommendations

Recurrence of cholestatic liver disease

Recurrent AIH, PBC and PSC vary between 10–50%; however, the impact on graft function and patient survival is minimal [[422], [423]]. Nevertheless, a recent study has shown that recurrent PSC may lead to graft loss in up to 25% of patients with recurrent disease [157]. In addition, the rate of recurrent PSC seems to be increased in living donor LT [424].

Recommendations

Management of HCC recurrence

Literature on the management of recurrent HCC after transplantation is very scarce. Most efforts have been placed in a good selection of candidates for transplantation in order to minimize HCC recurrence. The latter is associated with a ominous prognosis since therapeutic options at time of diagnosis are usually very low: HCC recurrence occurs in 8–20% of recipients and is usually seen during the first 2 years after LT, with a median survival lower than 1 year [83].

One of the main research topics in patients undergoing LT due to HCC is the effect of immunosuppression on HCC recurrence. There are no RCTs available to demonstrate that stronger immunosuppression is associated with a higher risk of recurrence. Regarding the potential impact of mTOR inhibitors on HCC recurrence, this is still a matter of debate. mTOR inhibitors have gained popularity in the transplantation context because of their low nephrotoxicity and potential anti-tumour effect. The mTOR pathway is a key regulator of cellular proliferation and angiogenesis implicated in carcinogenesis. SRL and EVR have been approved by the Food and Drug Administration for treatment of advanced renal cell carcinoma after failure of first-line treatment (sunitinib or sorafenib). Nevertheless, the only solid data showing an impact of mTOR inhibitors on HCC growth are based on preclinical models [425]. Clinical data suggesting a potential benefit rely on uncontrolled pilot and retrospective analyses [[83], [425], [426]]. Currently, mTOR inhibitors are been assessed in several clinical trials for the treatment of advanced HCC, and as adjuvant therapy in HCC patients after LT and TACE. Results of these trials will emerge in the coming years [425].

A large RCT in non-transplant patients demonstrated that systemic treatment with the multikinase inhibitor sorafenib prolonged survival in patients with advanced HCC [427]. Since most HCC recurrences after LT are associated with systemic tumour dissemination, a few retrospective cohort studies, isolated case reports and a small case-control study have assessed the safety and efficacy of sorafenib in this setting [[428], [429]]. Although the data suggest that sorafenib might be associated with a benefit in survival with an acceptable safety profile, a recommendation on its use cannot be established with the current data.

A different situation arises in patients who have progressed to liver cirrhosis over the years, in most cases due to hepatitis C recurrence. In the latter situation, de novo HCC may occur and treatment should probably follow the same algorithms used for immunocompetent patients: liver resection, radiofrequency ablation or TACE (when technically possible) and even retransplantation may be indicated in selected cases.

Recommendations

Management of renal dysfunction

The majority of patients who survive the first six months after LT then present with impaired kidney function. Between 30–80% of patients develop chronic kidney disease stage 3–4 with a cumulative risk of ESRD requiring maintenance dialysis or even renal transplantation of 5–9% within the first 10 years post-LT [[295], [430]]. The number of patients with renal failure after LT has recently further increased due to the implantation of MELD based allograft allocation and the need to use marginal grafts.

Chronic renal failure is a very important issue regarding the management of LT patients. Renal impairment may be present already before LT, may develop or be aggravated during LT and/or occur in the early and late post-operative course. The aetiology of impaired kidney function following LT is multifactorial, including (long-term) exposure to CNI-based immunosuppressive regimens, preoperative kidney dysfunction (hepatorenal syndrome, pre-existing kidney diseases), perioperative acute kidney injury and hypertension, diabetes mellitus, atherosclerosis pre- and/or post-LT. CNIs are considered to be responsible for >70% of cases of ESRD after LT [430]. Acute kidney injury as well as chronic renal disease are associated with a statistically significant increased risk of mortality in the early and late post-LT course [[295], [431]].

Therefore, a continuous screening for and sufficient treatment of potential risk factors as well as a regular monitoring of renal function and adjustment of the immunosuppression is mandatory. There is currently no guideline regarding the place of renal biopsy in the setting of kidney injury after LT [311]. Studies have been conducted with the aim either to prevent or to reduce CNI associated renal failure by using CNI-free immunosuppressive regimens or by early CNI minimization [[310], [321], [432]]. However, until now CNI-free regimens have been associated with a high rate of acute cellular rejection.

Recommendations

Prevention and treatment of infections

Infectious complications are a major cause of morbidity and mortality following transplantation and indeed, around 2/3 transplanted individuals will develop an infection after transplantation. Prevention of infections and an aggressive diagnostic strategy are cornerstones in solid organ transplant programs.

Antimicrobial prophylaxis has decreased the incidence and severity of post-transplant infections and has contributed to increased patient survival [433]. From a simplistic point of view one can divide the type of infections occurring after LT in three different timelines [434]: 1) first month after the procedure, where nosocomial infections mostly related to surgery and post-operative care are common; 2) 2–6 months after transplantation, when immunosuppression is at its maximum and opportunistic infections and reactivation of latent infections are the major cause of morbidity; and 3) later than 6 months after the procedure, when community-acquired infections are the major source of problems.

Bacterial infections

Bacterial pathogens are the most common causes of infection after LT. Gram-negative bacteria, such as Escherichia coli, Enterobacter, Pseudomonas are the most common in a majority of series. Bacterial infections involve mainly the surgical site, the abdominal cavity, the urinary tract and the bloodstream. Although surgical site infections are associated with an increase in morbidity rate, intra-abdominal infections are associated with increased mortality and graft loss [435].

Viral infections
CMV

CMV infection remains the most significant opportunistic infection in liver transplant recipients. An adequate prophylactic strategy has been shown to significantly reduce its incidence but it still produces relevant morbidity. The most common clinical syndromes are viremia, bone marrow suppression and involvement of the gastrointestinal tract (i.e. colitis) and the liver (hepatitis) [[436], [437]].

The use of CMV-seropositive donors in CMV-seronegative recipients increases the risk of developing CMV infection as well as past acute rejection episodes and the use of intense immunosuppression.

Treatment with ganciclovir or valganciclovir should be implemented in patients with persistent or increasing viremia (CMV infection), and in all individuals in whom CMV infection evolves into CMV disease. The detection of viremia by CMV-PCR during the first months after LT is essential for early diagnosis of this common infection [[433], [436], [437]]. Intravenous ganciclovir or oral valganciclovir is the treatment of choice in patients with mild disease, whereas intravenous ganciclovir should be used in patients with more severe infections [[436], [437]].

EBV

Patients with EBV seropositivity before LT, and patients treated with aggressive immunosuppressive regimens (i.e. anti-lymphocyte globulin) are at higher risk of developing post-transplant lymphoproliferative disorders (PTLD) [438]. PTLD should always be suspected in liver transplanted patients, especially those at high risk, presenting with fever, weight loss, night sweats, even in the absence of lymphoadenopathy. Radiographic analysis should be performed as EBV viremia is not a diagnostic for EBV-associated PTLD [439].

The first step in treating patients with PTLD is reducing the immunosuppressive therapy. Additional therapies including rituximab, chemotherapy, radiation and surgery may be necessary if no response is achieved by immunosuppression reduction. The multidisciplinary assessment, including oncologist, should always be performed.

HEV

Despite the prevalence of HEV infection in Central European, liver transplant recipients is low, it can result in graft hepatitis and graft dysfunction after LT. Therefore screening for HEV RNA should be part of the diagnostic work-up of patients who are evaluated for LT.

Fungal infections

Over the last two decades, the overall incidence of invasive fungal infections remained unchanged; however, a significant decline in the incidence of invasive candidiasis and an insignificant increase in invasive aspergillosis has been shown [440]. Identified risk factors for invasive fungal infections are: a decrease in the length of transplant operation, intraoperative transfusion requirements, cold ischaemic time, use of roux-en-Y biliary anastomosis, PVT, biopsy-proven rejection episodes, retransplantation and renal replacement therapy [[440], [441], [442]].

Diagnosis of invasive fungal infections is difficult since blood cultures are relatively insensitive. Other tests have a variable accuracy: beta-d-glucan (for Candida) and galactomannan testing (for Aspergillus) have inconsistent accuracy, whereas serum and cerebrospinal cryptococcal antigen testing is highly reliable [437]. Antifungal therapy relies not only on an adequate election of the drug but also on a reduction in immunosuppression.

Candida species

Fungemia or peritonitis due to Candida albicans and non-albicans Candida species (e.g. C. glabrata, C. krusei, C. tropicalis) are leading causes of early invasive infection after LT. Oral prophylaxis against Candida species is recommended during the first months, as it reduces mortality due to fungal infection. At present, fluconazole is the most commonly used antifungal agent [443].

Aspergillus

Infection with Aspergillus species may be activated in individual colonized pre-transplantation or as a result of new environmental or nosocomial exposures. The lungs are the primary site of infection, and dissemination commonly involves the central nervous system. Clinical signs of central nervous system infection necessitate radiologic and cerebrospinal fluid evaluations.

Prophylaxis against Aspergillus is only recommended in certain high risk situations: prolonged use of corticosteroids before transplantation (such as AIH), acute renal failure requiring hemodialysis, acute liver failure, retransplantation, high transfusion rate during surgery, early re-exploration after LT and maintained renal failure after LT. If the risk of infection is moderate inhaled amphotericin B is the treatment of choice, but if the risk is high (3 or more risk factors) micafungin is indicated [437].

Pneumocystis jirovecii

Pneumocystis pneumonia is rare during trimethoprim-sulphamethoxazole (TMP-SMX) prophylaxis [444]. Prophylaxis against Pneumocystis jiroveci is mainly accomplished by 6–12 months of cotrimoxazol (dapsone or pentamidine can be used if sulfonamide allergy) [[437], [444]]. The clinical presentation is insidious with shortness of breath occurring early but with relatively subtle findings by chest radiography. TMP-SMX is the agent of choice but may provoke renal toxicity. Corticosteroids are useful as adjunctive therapy to both reduce pulmonary inflammation and reduce post-infection fibrosis.

Mycobacteria

Active tuberculosis can be diagnosed in 0.47–2.3% of liver transplanted patients, and mostly in the first 12 months after LT [[445], [446]]. Fever, night sweats and weight loss are common symptoms; however, since extrapulmonary tuberculosis are present more frequently in liver transplanted patients compared to the general population, atypical presentations can occur.

Treatment of latent tuberculosis is relevant since diagnosis of this infection in transplant patients is not always easy and has a high mortality rate. Treatment with isoniazid for 9 months (supplemented with vitamin B6) is the standard therapy and should be indicated in the following situations: PPD positive skin test, history of untreated tuberculosis, chest radiography findings compatible with tuberculosis.

Treatment of active tuberculosis in liver transplant recipients is not standardized and it is not based on RCTs [447]. Moreover, active tuberculosis therapy is complicated by the interactions between antituberculous and immunosuppressive drugs, and by the potential hepatotoxicity associated with first-line tuberculosis treatment [445]. Therefore, in cases of non-severe tuberculosis, treatment should include isoniazid and ethambutol avoiding rifamycins. Levofloxacin can replace isoniazid if its use is not possible. Patients with severe tuberculosis should be treated with rifamycin during the initial and maintenance phases.

Recommendations

Prevention and treatment of diabetes, hypertension, cardiovascular disease (metabolic syndrome), bone disease and de novo tumours

Metabolic syndrome

Metabolic syndrome is a mounting challenge in the management of LT recipients. The clinical features of metabolic syndrome, in particular insulin-resistant (type 2) diabetes mellitus, obesity, dyslipidaemia and arterial hypertension, either alone or in combination contribute to late post-operative morbidity and mortality. The prevalence of metabolic syndrome lies between 50–60% in the LT population [420]. Diabetes mellitus is diagnosed in 10–64% of LT patients, obesity (BMI >30 kg/m2) in 24–64%, dyslipidaemia in 40–66% and arterial hypertension in 40–85% [437].

Due to the high prevalence of metabolic syndrome and its different clinical features, LT recipients have a significantly increased risk of cardiovascular events and mortality compared to an age and gender-matched general population [448]. Based on several publications this elevated risk of cardiovascular diseases ranges from around 10% at five years to up to 25% at 10 years [[448], [449]]. Therefore, cardiovascular disease accounts for almost a quarter of deaths in the long-term follow-up after LT [[449], [450]].

Numerous publications have shown that the currently issued immunosuppressive regimens cause both an exacerbation of pre-existing systemic and metabolic disorders and de novo post-LT arterial hypertension, hyperlipidaemia, diabetes and obesity [449].

Therefore, a continuous cardiovascular risk stratification and an aggressive management of the metabolic syndrome, in particular, the rapid detection and treatment of metabolic disorders, as well as modification of risk factors including tailoring the immunosuppressive regimen are mandatory in order to avoid cardiovascular morbidity and mortality.

In patients treated with 3-hydroxy-3-methyl-glutaryl coenzyme A (HMG-CoA) reductase inhibitors, potential interactions with CNIs should always be considered, due to the fact that both statins and CNIs are metabolized by cytochrome P450–3A4. This can result in increased statin concentrations, with an increased risk of developing rhabdomyolysis. Therefore statins should always be started at a lower dose and gradually titrate upwards, and patients should be followed-up closely to detect any potential side effects.

Hydrophilic statins such as fluvastatin and pravastatin are preferred as they are not metabolized by cytochrome P450–3A4 and they may cause less metabolic interactions.

Recommendations

Bone disease

Patients with end-stage liver disease present with decreased bone density compared with age-matched control population. Bone loss accelerates in the first 6 months after LT, independently of the pre-transplant bone mineral density, and it is associated with increased risk of fractures causing pronounced morbidity and reduced QoL [[451], [452]]. The first 6–12 months after LT, bone loss reverses and there is a gain in bone density.

Among risk factors for developing post-transplant bone disease the most important is a low bone mineral density before LT [[453], [454]]. This can be caused, in general, by malnutrition and physical inactivity, by malabsorption of vitamin D in cholestatic liver disease, steroid use in patients with AIH and direct toxicity in alcoholic patients [455]. Post-LT immunosuppression regimen, in particular steroids, female sex, older age, lower BMI and renal dysfunction represent risk factors for low bone mineral density and an increased incidence of fractures.

Therefore, a regular measurement of bone mineral density is recommended pre- and post-LT. In the case of osteopenia and low bone mineral density, calcium and vitamin D supplementation and, if tolerable preoperative, a weight-bearing exercise should be started. Bisphosphonate therapy must be considered for patients with osteoporosis and/or recurrent fractures.

Recommendations

De novo malignancies

Besides cardiovascular diseases de novo malignancies are the leading cause of mortality following the first post-LT year. Observational studies have shown a 2–3-fold elevated risk of solid organ cancers and a 30-fold or higher increase in the rate of lymphoproliferative malignancies compared to the general population [[450], [456], [457]]. Several papers have reported an incidence of de novo cancers ranging from 3% to 26%, mainly dependent on follow-up duration, with a continuous rise in risk up to 19% and 34% at 10 and 15 years, respectively, following LT [[450], [456], [457]].

The major cause of de novo malignancies in the post-LT course is related to the loss of immunovigilance induced by immunosuppressive agents, as well as other risk factors associated with carcinogenesis, such as viral infections with oncogenic potential (e.g. EBV, human papilloma virus), PSC, smoking and alcohol abuse. In general, an increased frequency is not detected in many of the common cancers in the absence of identified risk factors.

Skin cancer is the most common de novo malignancy in patients who underwent LT [458]. Among these, non-melanoma skin cancers such as squamous and basal cell carcinomas are more frequent than melanomas. Their incidence is 20-fold higher in liver transplant recipients compared to age and sex-matched population, and generally tend to be more aggressive, recurring and metastatizing more frequently than in non-transplant population [459]. Major risk factors for developing non-melanoma skin cancers after LT include: older age, chronic sun exposure and sunburn, fair skin, and a history of skin cancer [460].

Patients with alcoholic cirrhosis are of particularly increased risk for the development of cancer in the upper gastrointestinal, oropharyngeal-laryngeal, as well as lung cancers [[450], [461]]. A positive smoking history both pre- and post-LT further increases the risk of head/neck and pulmonary de novo malignancies in these patients underscoring the importance of discontinuing smoking in LT candidates and recipients [462].

Patients with EBV seropositivity before LT, and patients treated with more aggressive immunosuppressive regimens (i.e. anti-lymphocyte globulin) are at a higher risk of developing PTLD. Therefore PTLD should always be suspected in liver transplanted recipients, especially those at high risk, who present with fever, weight loss and night sweats, even in the absence of lymphoadenopathy.

Significantly higher rates of colorectal cancer have been demonstrated for patients with PSC and inflammatory bowel disease in the post-LT course [450]. Therefore, annual screening colonoscopies are recommended in these patients [463].

The development of de novo solid organ cancers has a major impact on the outcome of LT due to a poor prognosis in the majority of patients with de novo neoplasia. The probability of survival for LT recipients after the diagnosis of de novo cancers mainly depends on tumour location, type and stage. In general, the outcome is worse compared to the general population with the same malignant diseases. One recent study showed a median survival lower than 3 years after the diagnosis of de novo cancer [457].

Many known risk factors for de novo malignancies cannot be modified, such as age and underlying liver disease. Therefore, regular cancer surveillance programs have been proposed by several groups; however, none of these recommendations are based on scientific evidence [463]. A recent paper has shown improvements of both cancer detection rates and non-cutaneous cancer patient survival after applying a strict surveillance protocol to all LT recipients [457]. More data, however, are needed to define the optimal surveillance protocol after LT with individualized emphasis laid on patients’ particular risk profiles.

Recommendations