Extracorporeal therapies for post-liver transplant recipient: The road less traveled

Abstract

Extracorporeal therapies have a definite role in patients with acute liver failure, acute on-chronic liver failure, and progressive chronic liver disease. They act as a bridge-to-transplant in these patients. With the increasing success of liver transplantation, the immediate postoperative complication spectrum continues to expand. Extracorporeal therapies can play an important role in managing these complications. However, the literature on extracorporeal therapies in the post-liver transplant period is limited. This review article discussed various extracorporeal therapies that are still evolving or marred by limited evidence but can improve patient outcomes. These extracorporeal therapies can be divided into two subgroups: (1) Therapies for infective complications. Endotoxin and cytokine adsorption columns; and (2) Therapies for noninfective complications like small for size syndrome, primary allograft nonfunction, early allograft dysfunction, hyperacute rejection, hepatopulmonary syndrome, etc. (plasma exchange, double plasma molecular adsorption, molecular adsorbent recirculation system, and extracorporeal membrane oxygenation, among others).

Key Words: Critical care; Intensive care unit; Liver transplant; Extracorporeal therapies; Polymyxin B hemoperfusion; Cytokine adsorption; Therapeutic plasma exchange; Molecular adsorbent recirculation system; Extracorporeal membrane oxygenation

Core Tip: With the growing number of liver transplants, both infective and noninfective complications are bound to grow. The discussion on extracorporeal therapies sheds light on innovative approaches that may improve patient outcomes, showcasing advancements in medical technology. By acknowledging the limited literature and ongoing research, the article highlighted the need for further studies to validate the efficacy and safety of these emerging therapies, which is crucial for informed clinical practice.

INTRODUCTION

Liver transplantation (LT) is now the ultimate treatment option for patients with end-stage liver disease and for many who suffer from acute liver failure. The United Network for organ-sharing data reported that for the first time more than 10000 LTs were performed in 2023 in the United States. With the growing frequency of transplants, complications are inevitable, thereby deeming the role of extracorporeal therapies vital. However, most of these extracorporeal therapies are either still evolving or marred by limited literature and evidence.

The extracorporeal therapies for a patient after LT can be divided into those for infective complications and noninfective complications (Figure 1). The principal mechanism of extracorporeal therapies is described briefly in Table 1[1-7]. Although the effectiveness of modern immunosuppressive drugs has enhanced both graft and patient survival following LT, it has also increased the rate of infections. More than 50% of LT recipients experience infections post-transplant[8]. Over the years, an increased risk of infections from multidrug resistant organisms (MDRO) has been observed. Prolonged use of preoperative broad-spectrum antibiotics, surgical re-explorations, and high model for end-stage disease score (> 25) are associated with an elevated risk of infection from MDRO post-LT[9,10]. These MDROs are resistant to most antibiotics and combinations, making treatment challenging. As a result, timely diagnosis and prompt initiation of appropriate treatments are essential to improving patient outcomes. In this scenario, extracorporeal therapies play a significant role in modulating the sepsis response and improving survival.

Figure 1 Proposed extracorporeal therapies for patients after liver transplantation. DPMAS: Double plasma molecular adsorption system; EAD: Early allograft dysfunction; ECMO: Extracorporeal membrane oxygenation; HPS: Hepatopulmonary syndrome; MARS: Molecular adsorbent recirculation system; PANF: Primary allograft nonfunction; PPS: Portopulmonary syndrome.

Table 1 Principal mechanism of extracorporeal therapies.

Modality	Mechanism
PMX hemoperfusion	Endotoxin, a cell wall component of gram-negative bacilli, is primary reason for generating inflammatory mediators, the emergence of septic shock, and multiple organ failure
	PMX, a polycationic antibiotic, attaches to lipid A (part of endotoxin), thus neutralizing it
	Toraymyxin® (Toray Medical Co., Ltd., Japan) is a PMX-immobilized fiber blood-purification column
	PMX is immobilized covalently on the surface of polystyrene-derived, polypropylene-reinforced conjugated carrier fiber for selective endotoxin adsorption from the blood
Cytokine adsorption (Cytosorb®)	Cytokine storms are caused by a high number of circulating cytokines and immune cell hyperactivation
	One of the most difficult problems in liver transplantation is adverse immunological reactions mediated by cytokines against the allograft, which can cause early malfunction, severe rejection, and chronic damage
	Hemoperfusion cartridge containing polymer beads to adsorb proinflammatory and anti-inflammatory cytokines but not endotoxins
Oxiris® hemofilter (cytokine + endotoxin removal)	During manufacture, 4500 UI/m2 heparin is pregrafted onto the oXiris® membrane
	A layer of PEI, a positively charged molecule that promotes greater biocompatibility, constitutes the membrane surface treatment
	PEI grafting has the capacity to adsorb large negatively charged molecules, such as endotoxins, due to its higher concentration of free amino groups, which carry a positive charge
	As a result, the oXiris® membrane is composed of three distinct layers
	Thanks to this innovative design, one device may perform four functions: Renal support; cytokine removal; endotoxin removal; and local anticoagulant treatment
Therapeutic plasma exchange	Elimination of large molecules from the blood, such as albumin-bound and water-soluble toxins, followed by replacement with plasma and/or albumin
	Toxins removed include cytokines, endotoxins, bilirubin, bile acids, ammonia, and aromatic amino acids
Molecular adsorbent recirculation system	The system utilizes a dialysate enriched with albumin to aid in the removal of albumin-bound toxins
	It consists of three separate fluid compartments: A blood circuit, a circuit with 600 mL of 20% human albumin that includes a charcoal column and an anion exchange resin column, and a dialysate circuit
DPMAS	Uses ion exchange resin and neutral macroporous resin to create a new artificial liver model, which may have a better adsorption effect on inflammatory mediators and bilirubin
	It also reduces the risk of allergic reactions and transmitted diseases, among other risks by avoiding any replacement with plasma or albumin
	DPMAS also has drawbacks, like the inability to supplement albumin and coagulation factors
DIALIVE	DIALIVE uses a dual filtration system coupled in series with a renal dialysis machine (Prismaflex, Baxter)
	The first filter is made up of a membrane that enables albumin and cytokines to be ultrafiltered (Septex, Baxter, United States); the second filter, called oXiris® (Baxter, United States), adsorbs damage-associated molecular patterns and pathogen-associated molecular patterns
	20% bottled albumin is added in equal amounts to replace the lost albumin
Prometheus	The Prometheus system operates based on the principle of plasma separation using a filter that allows albumin, clotting factors, and fibrinogen to pass through
	It requires an albumin filter (AlbuFlow® AF 01), a neutral resin adsorber (prometh® 01), an adsorber for anion exchange (prometh® 02), and a polysulfone® high-flux dialyzer with the corresponding tubing system
	Endogenous albumin is passed through the circuit using an Albuflow filter
	Albumin is reactivated in prometh 01 and 02 adsorbers and returned to circulation
	Blood then passes through the polysulfone filter, where it is treated by conventional high flux hemodialysis, eventually returning blood to the patient
ECMO	In veno-venous ECMO, deoxygenated blood is drawn from the venous catheter and pumped into an oxygenator
	This device functions like a synthetic lung, facilitating the exchange of gases by removing carbon dioxide and adding oxygen
	Inside the oxygenator, air and oxygen circulate through small, hollow fibers
	As blood flows through these fibers, oxygen is transferred into the red blood cells, while carbon dioxide is expelled into the fibers
	The carbon dioxide is then eliminated through the exhaust, and the oxygenated blood is returned to the patient via the catheter
	In veno-arterial ECMO, blood is drained from the venous side and pumped into the arterial side, bypassing the heart

DPMAS: Double plasma molecular adsorption system; ECMO: Extracorporeal membrane oxygenation; PEI: Polyethyleneimine; PMX: polymyxin B.

In addition to infective complications, post-LT recipients may also encounter immediate noninfective complications, such as primary allograft nonfunction (PANF) right after reperfusion and early allograft dysfunction (EAD) shortly after transplantation or small for size graft. Graft dysfunction often results in several postoperative complications, high mortality, and an increased risk of organ loss[11,12]. Thus, extracorporeal therapies can play a bridging role in supporting the transplanted organ along with conventional management.

This review discussed the literature on the role of extracorporeal membrane oxygenation (ECMO), especially for hepatopulmonary syndrome (HPS) and portopulmonary syndrome in the immediate postoperative period. Herein, a comprehensive literature search was conducted on articles published across the following databases: (1) PubMed; (2) PubMed Central; (3) Cochrane; (4) Elsevier Clinical Key; and (5) Google Scholar. The keyword search included LT, extracorporeal therapies, polymyxin (PMX) B hemoperfusion, cytokine adsorption, EAD, PANF, small for size graft, therapeutic plasma exchange, molecular adsorbent recirculation system, double plasma molecular adsorption system (DPMAS), HPS, and veno-venous ECMO (VV ECMO). The inclusion criteria were publications on post-LT adults (> 18 years) and in the English language, whereas studies conducted in the pediatric population and in languages other than English were excluded.

EXTRACORPOREAL THERAPIES FOR INFECTIVE COMPLICATIONS

PMX adsorption column

Patients with end-stage liver disease may benefit greatly from LT, which has a 90% 1-year and 80% 5-year survival rate[13]. One of the major difficulties following LT is managing infections, as the factors contributing to post-transplant infection risk are complex and challenging to control[14]. Hepatic abscesses, cholangitis, infected fluid collections, or peritonitis (intraabdominal infections) account for about half of the early infections post-transplant. Septicemia, pulmonary infection, and wound infection are 34%, 31%, and 10%, respectively[15-17]. Re-transplantation and transplant failure are linked to intraabdominal infections[18,19]. Approximately 67% of infections in the early stages after a transplant are caused by gram-negative bacteria[15,20]. With an estimated frequency of 14% and a fatality rate of up to 54%, septic shock is a significant cause of morbidity and mortality following solid-organ transplantation[21].

PMX B has been shown to protect against endotoxemia and septicemia as assessed in animal models[22-24]. The Early Use of PMX Hemoperfusion in Abdominal Sepsis (EUPHAS) trial was a prospective, multicenter, randomized controlled trial (RCT) that was carried out at ten Italian tertiary care intensive care facilities. A total of 30 patients received conventional therapy, while 34 patients received conventional therapy plus two sessions of PMX B hemoperfusion. The findings revealed that when combined with standard therapy, PMX B hemoperfusion considerably improved organ dysfunction (change in sequential organ failure assessment score, -3.4 vs -0.1; P value < 0.001) and hemodynamics (76–84 mmHg; P value = 0.001). In a targeted population suffering from intraabdominal gram-negative infections that caused septic shock or severe sepsis, it also decreased 28-day mortality [unadjusted hazard ratio (HR) = 0.43; 95% confidence interval (CI): 0.20–0.94; adjusted HR = 0.36; 95%CI: 0.16–0.80][25].

Payen et al[26] focused on the early use of PMX B hemoperfusion in patients with septic shock due to peritonitis, including those requiring urgent surgery for visually confirmed peritonitis. Compared to conventional therapy for peritonitis-induced septic shock, this multicenter RCT showed no significant change in mortality (19.5% vs 27.7%; P value = 0.14), and PMX B hemoperfusion did not lead to any improvement in organ failure (P value = 0.78).

In the EUPHRATES study, Dellinger et al[27] assessed the impact of targeted PMX B hemoperfusion on 28-day mortality in patients with septic shock with high endotoxin levels. A total of 449 critically ill adult patients with septic shock and an endotoxin activity (EA) assay result of 0.60 or higher participated in this multicenter RCT. They inferred that compared to sham treatment with standard medical care, PMX B hemoperfusion treatment did not reduce the 28-day mortality (34.5% vs 37.7%, P value = 0.49) in individuals with septic shock and high EA.

Klein et al[28] conducted a post-hoc analysis of the EUPHRATES trial to examine the effects of PMX B hemoperfusion usage in patients with an EA ranging from 0.60-0.89 and septic shock. Based on post-hoc analysis, they proposed quantifiable responses to variations in mean arterial pressure (P value = 0.04), ventilator-free days (P value = 0.004), and 28-day mortality in patients with septic shock and an EA > 0.60-0.89 (HR = 0.56, 95%CI: 0.33-0.95, P value = 0.03).

Nonetheless, these trials all had limitations. In the Early Use of PMX Hemoperfusion in Abdominal Sepsis trial, there were only 64 patients, and it was terminated early with a considerable risk of type I error. Although 232 patients participated in the ABDOMIX trial, the patients were less sick, and 11% of patients had incomplete PMX B hemoperfusion. Moreover, neither of the two studies had a limitation of not measuring endotoxin levels. In the EUPHRATES trial, the post hoc analysis set an arbitrary cutoff for the endotoxin levels. Another study suggested that these findings should not be taken as evidence of therapeutic efficacy but interpreted as hypothesis generating[29]. Surviving sepsis guidelines 2021 also did not recommend the use of PMX B hemoperfusion[30]. Nonetheless, cartridges with a larger surface area in contact with blood reduce the possibility of system saturation in the presence of high levels of endotoxins. The next step for this therapy might be defining a customized dose of PMX B hemoperfusion and the earliest appropriate application for patients with elevated EA in cases of high inoculum.

The current literature on the use of PMX B hemoperfusion is very limited. A study on the safety of PMX B hemoperfusion in recipients of kidney transplantation and LT recruited 28 (39.5%) patients with an EA > 0.60 to receive the treatment at a blood flow rate of 100 mL/minute for 2 h. PMX B hemoperfusion was administered to each patient until their EA was < 0.40. Consequently, the inflammatory and hemodynamic frameworks were stabilized. All patients were still alive and had minimal levels of EA and satisfactory graft function at the 30-day follow-up. The study concluded that EA had the ability to identify patients who could benefit from targeted treatment. It proved valuable in determining therapeutic dosages by accurately assessing the effects of PMX B hemoperfusion therapy. Additionally, it detected endotoxemia even in the absence of culture evidence of gram-negative infection[31].

In another study, Ruberto et al[1] detected gram-negative bacteria in 10 patients who underwent orthotopic LTs and developed sepsis/septic shock. PMX B hemoperfusion was performed three times in each patient, but no adverse event was reported. The mean arterial pressure improved from 64 mmHg ± 5 mmHg to 89 mmHg ± 4 mmHg between the baseline and the third treatment; in addition, the doses of norepinephrine and dobutamine were lowered from 1.3 µg/kg/minute to 0.001 µg/kg/minute and from 6.4 µg/kg/minute to 1 µg/kg/minute, respectively, whereas the PaO₂ to FiO₂ ratio rose from 214 mmHg to 291 mmHg.

Sato et al[32] evaluated the impact of hemoperfusion using a PMX B-immobilized fiber column on liver function following ischemic reperfusion injury. A total of 11 pigs were subjected to 120 min of total hepatic vascular exclusion and then monitored for 360 min. The rate pressure product (heart rate × end-systolic arterial blood pressure) and hepatic tissue blood flow were notably high in the PMX B hemoperfusion group. Compared to the control group, the portal venous blood flow in the PMX B hemoperfusion group differed significantly (P value < 0.05) at 360 min after reperfusion and was maintained at about 70% of the flow before ischemia. The serum aspartate aminotransferase levels in both groups increased progressively post-reperfusion; however, after 360 min, the enzyme levels in the PMX B hemoperfusion group were considerably lower (P value < 0.05). In this pig model, PMX B hemoperfusion treatment decreased the hepatic warm ischemia-reperfusion injury produced by total hepatic vascular exclusion.

Recently, a retrospective, multicenter, observational PMX-OTIC study evaluated the optimal time for initiation of PMX B hemoperfusion. A total of 48 patients were analyzed. The results showed that in terms of the timing of PMX B hemoperfusion initiation relative to the noradrenaline equivalent dose peak, the group initiated between -4 hours and 4 hours had the highest survival rate (229/304, 75.3%) compared to those starting ≤ -4 hours (50/75, 66.7%) and ≥ 4 hours (66/101, 65.4%) (P value = 0.085). In multivariable regression analysis, -4 hours to 4 hours was used as the reference and initiation at ≤ -4 hours was found to be associated with an odds ratio of 1.96 (1.07–3.58, P-value = 0.029) for 28-day mortality, while initiation at ≥ 4 hours showed an odds ratio of 1.64 (0.94–2.87, P value = 0.082). Hence, PMX B hemoperfusion induction at the time of the peak of vasopressor dose effectuated low mortality[33].

Thus, PMX B hemoperfusion may play a crucial role in patients experiencing gram-negative sepsis or septic shock after LT in improving the hemodynamic status and probably survival. These effects need to be substantiated in larger and better-planned trials, considering the timing of initiation, endotoxin levels, their clearance, and the number of PMX columns needed for endotoxin clearance. Until then, the limited literature and anecdotal evidence may be the best bet for these patients.

Cytokine adsorption column

The in vitro studies have revealed clearance rates of > 90%-95% of cytokines using CytoSorb^®[34]. Nevertheless, clinical research is still rare and is frequently restricted to case series that present positive findings for blood lactate levels and hemodynamic indices[35]. Despite considerable clearance during sessions, a recent RCT comparing standard therapy to hemoperfusion with CytoSorb^® (6 h per day for 7 days) showed no decrease in interleukin-6 plasma levels over time[36].

Another hemoperfusion cartridge commonly used for cytokine and endotoxin removal is oXiris^®. In a case report on septic shock after LT, Li et al[37] successfully stabilized the patient with continuous renal replacement therapy (oXiris^® hemofilter) and antibiotics.

In a literature review on the clinical utility of extracorporeal cytokine hemadsorption therapy, Bonavia et al[38] concluded that extracorporeal blood purification using cytokine hemadsorption techniques decreased cytokine levels in patients with sepsis and septic shock; however, these benefits have not translated into consistent clinical improvement.

Gaspari et al[39] conducted a retrospective analysis on 7 patients assessing blood purification in hepatic dysfunction after LT or extensive hepatectomy and did not find any potential advantages of CytoSorb^® in restoring patients’ clinical and laboratory parameters to a healthy state.

The cytokine storm and cytokine release syndrome represent complex, potentially fatal inflammatory responses that can occur following various triggers, including infections, therapies, and LT. These conditions, characterized by hyperactivation of the immune system and elevated cytokine levels, pose significant challenges in patients undergoing LT and often lead to graft malfunction, severe rejection, or chronic damage.

While extracorporeal blood purification techniques such as CytoSorb^® and oXiris^® seem promising in adsorbing proinflammatory mediators and improving certain clinical parameters, the clinical evidence remains inconclusive. Despite some positive in vitro findings and case reports indicating beneficial effects on lactate levels, hemodynamics, and septic shock stabilization, the translation of these results into consistent clinical improvement is still lacking. Thus, larger RCTs are needed to definitively establish the safety, efficacy, and clinical benefits of these technologies in managing cytokine-driven complications in LT. Until then, cytokine hemadsorption therapies should be cautiously approached as their role in clinical practice remains uncertain.

EXTRACORPOREAL THERAPIES FOR NONINFECTIVE COMPLICATIONS: THERAPIES FOR PANF, SMALL FOR SIZE GRAFTS, AND EAD

Small for size graft is defined by a graft-to-recipient body weight ratio of < 0.8 for living-related LT[40,41]. Acute allograft rejection is still a frequent side effect of LT, accounting for 20%-40% of cases, despite recent advancements in immunosuppressive medication; rejection usually occurs within a month following LT[42,43]. About 5%-10% of LTs result in PANF[44,45]. Certain donor features, including advanced age, steatosis, and prolonged ischemia periods, seem to contribute to the development of graft dysfunction, even if the pathophysiological foundation for PANF remains unclear[46-48]. Shortly after transplantation, a diagnosis is made to rule out the alternative causes of acute graft malfunction, such as vascular thrombosis, hyperacute rejection, or accelerated acute rejection. Currently, immediate re-transplantation has been the standard of care for PANF.

Therapeutic plasma exchange

After LT, therapeutic plasma exchange (TPE) has proven to be a crucial adjuvant in treating immune-mediated causes of graft malfunction. For instance, TPE has been used to effectively treat refractory hepatic allograft rejection and ABO-incompatible LTs. In a study by Mandal et al[49], 5 patients diagnosed with PANF underwent TPE. All patients had a restored liver function except 1 patient who died of pulmonary embolism; the other 4 patients who survived had a functional allograft after 1 year. In contrast, 2 patients who underwent re-transplantation for PANF died within 3 months of the transplant.

Evidence for TPE in post-LT small for size graft and allograft rejection is collected primarily from case series and case reports. Khoo et al[50] reported a case series involving 3 patients: 2 patients had small for size syndrome; and 1 patient had biopsy-proven allograft rejection. The number of TPE sessions ranged from one to three, and all patients showed a substantial decrease in bilirubin levels along with evidence of graft recovery.

In a retrospective analysis, Choe et al[51] used TPE in EAD after LT evaluation. The study included 107 patients with EAD who underwent TPE compared to 36 patients with EAD who did not receive TPE over 13 years. EAD was defined as persistent hyperbilirubinemia (approximately 10 mg/dL) without biliary problems within 30 days after LT. The survival rates for the non-TPE group were 58.3% at 1 month and 22.2% at 1 year compared to the TPE group, which showed significantly better survival rates, with 82.2% at 1 month and 53.8% at 1 year. After the last TPE session, the TPE group also demonstrated a statistically significant decrease in both international normalized ratio (INR) and total bilirubin (P value < 0.05). A better prognosis was observed in patients who were responders to TPE and met certain criteria, such as age < 51 years, total bilirubin < 11.1 mg/dL, or INR < 1.15 following the final TPE. Moreover, TPE was associated with a reduced risk of death, while higher INR at EAD onset, male gender, and older age were associated with an increased risk of mortality.

In another retrospective analysis of 67 patients with EAD, 64% were treated with TPE. In the TPE group, 13 (45%) patients were deceased (P value < 0.001), but no deaths occurred in the non-TPE group within 90 days. The TPE group required more postoperative renal replacement therapy and had more complications related to sepsis than the non-TPE group. The authors concluded that the limitations of the retrospective investigation did not allow them to confirm whether TPE contributed to increased premature mortality[52].

In ABO-incompatible LTs, TPE was used to reduce the isoagglutinin titers by using fresh frozen AB plasma. The targeted isoagglutinin titer range was 1:8-1:16[53,54]. TPE sessions may vary depending on the center and the patient’s specific needs. Importantly, TPE is the only technique that can mechanically reduce isoagglutinin titers after transplantation; however, titer rebound requires additional TPE sessions.

Hence, TPE has demonstrated promising results as an adjunctive treatment for immune-mediated complications following LT, particularly in cases of allograft rejection, ABO-incompatible LTs and EAD. While case series and retrospective studies support its efficacy, the lack of large-scale RCTs limits definitive conclusions regarding its widespread use. The clinical response to TPE appears to vary depending on factors such as the patient’s age, bilirubin levels, and INR, underscoring the importance of a customized approach. Although TPE can significantly improve graft function and survival in certain cases, additional studies are required to define its role better and establish standardized protocols for its application in LT.

Other liver support devices

In a pilot study on the use of molecular adsorbent recirculation system (MARS) for EAD, the criteria for EAD were as follows: (1) Prothrombin time < 40%; (2) Aspartate aminotransferase or alanine aminotransferase > 1000 U/L; (3) Plasma disappearance rate of indocyanine green < 10%/minute within 72 h of reperfusion; and (4) Serum bilirubin > 10 mg/dL. Following MARS treatment, a substantial increase was noted in the plasma disappearance rate of indocyanine green, whereas a significant decrease was detected in serum bilirubin (P value = 0.002), serum creatinine (P value = 0.006), and aspartate aminotransferase (P value = 0.005). On the other hand, platelet count, albumin level, and prothrombin time were constant. Renal, neurological, and mean arterial pressure improvements were noted over time, and MARS therapy offered a secure method of treating EAD without negative consequences[55].

In another study involving 5 patients with EAD and 2 patients with PANF, MARS decreased serum bilirubin levels, bile acids, serum creatinine, and ammonia levels. Four of the five patients with graft dysfunction showed a significant improvement in hepatic encephalopathy and renal and synthetic liver functions, whereas patients with PANF did not show a similar improvement[56].

A retrospective review of catastrophic graft failure (massive exsanguination and liver fracture, portal vein and hepatic artery thrombosis, idiopathic necrosis, necrosis from inadequate donor flushing/primary nonfunction) concluded that timely hepatectomy combined with a portacaval shunt and re-transplantation could result in satisfactory outcomes, especially for patients receiving MARS support[57].

Lee et al[58] compared 15 patients who received 41 sessions of MARS with 16 patients who received 105 sessions of plasmapheresis. The study concluded that both therapies significantly decreased bilirubin levels with a similar overall survival rate.

The literature on the use of DPMAS as a bridge to transplant and bridge to recovery is evolving. However, evidence for its use in patients after transplant is virtually nonexistent. The principle of DPMAS is promising, rendering it an effective option for treating the above-mentioned complications. Nonetheless, additional evidence and knowledge are required before its clinical application in patients after LT.

The extracorporeal liver dialysis machine, DIALIVE, was designed to address the pathophysiological abnormalities that lead to the development of acute on-chronic liver failure (ACLF). DIALIVE may benefit patients in the following ways: (1) In addition to being dysfunctional, the circulating albumin in ACLF can trigger inflammatory responses; and (2) The buildup of damage-associated molecular patterns and pathogen-associated molecular patterns induces organ dysfunction and increases the risk of infections, ultimately leading to systemic inflammation. Thus, DIALIVE seeks to eliminate damage-associated molecular patterns and pathogen-associated molecular patterns and replace albumin[59], albeit the modality is still under investigation.

Prometheus was first described by Falkenhagen et al[60] in 1999. It is based on a combined membrane and adsorbent blood purification technique. Its unique contribution was the development of a polysulfone hollow-fiber filter with a sieving coefficient of 0.89 for human serum albumin but only 0.19 for fibrinogen and 0 for IgM. The HELIOS study, a multicenter RCT comparing Prometheus to standard medical therapy did not find any statistically significant survival benefit in patients with ACLF[61]. Hitherto, various studies have shown MARS to have a better outcome in terms of hemodynamic parameters compared to Prometheus[62,63].

Sentürk et al[64] studied Prometheus in patients with acute liver failure and ACLF and found improvement in biochemical parameters in hepatic encephalopathy. Komardina et al[65] showed hemodynamic and biochemical improvement with Prometheus in patients with alkaline phosphatase but without any difference in survival outcome. Thus, varying conclusions have been derived regarding the clinical benefit of Prometheus. Similar to DPMAS and DIALIVE, Prometheus is also based on sound principles, and its use in post-LT complications should be explored in future trials.

EXTRACORPOREAL THERAPIES FOR NONINFECTIVE COMPLICATIONS: HPS AND PORTOPULMONARY SYNDROME

ECMO

HPS linked to liver disease has emerged as an indication for ECMO. VV ECMO is mainly utilized for severe, acute, reversible respiratory failure after LT, with HPS being the most common etiology. Patients with HPS may undergo excessive pulmonary vasoconstriction, increased ventilation-perfusion mismatch, and hypoxemia due to vasoconstrictors released after LT. ECMO may preserve oxygenation until the patient’s intrapulmonary shunting improves[66].

Several case reports have presented that VV ECMO was utilized to salvage patients who developed severe hypoxia because of HPS after LT[67-69]. In a systematic review of the literature on ECMO for HPS, Wu et al[70] evaluated 16 studies from across four continents on 17 patients who were started on ECMO before, during, or after LT. The median interval between LT and the start of ECMO was 7 days [interquartile range (IQR) = 3-12 days]. The average duration of ECMO treatment was 13 days (IQR = 10-29 days). The average length of stay in the postoperative critical care unit was 40 days (IQR = 37-61 days), whereas the average length of stay in the hospital was 59.5 days (IQR = 42-77 days); 14/17 (82.4%) patients were discharged. Thus, the study concluded that ECMO was a viable option for patients with HPS undergoing LT and seemed to present better results than other reasons for cardiopulmonary failure in LT patients. Eventually, ECMO might play a major role in perioperative care for patients with severe HPS.

Patel et al[71] conducted a single-center, retrospective study of 29 patients who needed peri-LT ECMO support. In the study, 13 patients were supported with ECMO post-LT. Of these, 9 patients required VV ECMO for indications like acute respiratory distress syndrome or aspiration pneumonia, with a median duration of 17 days (IQR = 5-41 days) from LT to ECMO support. All except 1 patient survived at hospital discharge. Veno-arterial ECMO was required for extracorporeal cardiopulmonary resuscitation in 3 patients, while 1 patient required it for portopulmonary hypertension. For veno-arterial ECMO, the median time to initiation after LT was 3.5 days (IQR = 3-8 days). Subsequently, 3 patients were weaned and discharged from the hospital, while 1 patient was deceased due to multiorgan failure. Considering ECMO support for patients after LT presents significant challenges. However, promising outcomes have been reported globally, suggesting potential benefits of ECMO in carefully selected cases. Taken together, the current evidence for the use of VV ECMO in patients with HPS is limited to retrospective data, case reports ,and anecdotal evidence, and no studies have yet been planned in this area of concern in patients after LT.

CONCLUSION

The landscape of LT is evolving, particularly with the significant increase in procedures performed annually. While this progress offers hope to patients with end-stage liver disease, it also brings forth the challenge of managing infectious and noninfectious complications. Extracorporeal therapies have emerged as a promising avenue to address these complications, ranging from infections like gram-negative sepsis to conditions such as PANF and HPS. Table 2 outlines the advantages and disadvantages of various major extracorporeal therapies in post-LT complications. However, the current evidence supporting these therapies is still in the nascent stage, with many treatments characterized by limited literature and ongoing trials. Although previous results have shown potential benefits, they underscore the need for larger, well-designed RCTs to establish efficacy and safety. Until then, clinicians must navigate the uncertainties and carefully tailor treatment strategies for each patient based on individual risk factors and cost-benefit analysis of the emerging evidence. The journey towards optimizing post-transplant management is ongoing, and continued innovation in this field holds promise for improved survival and quality of life for LT recipients.

Table 2 Advantages and disadvantages of extracorporeal therapies in post-liver transplant care.

Modality	Advantages	Disadvantages
Polymyxin B hemoperfusion	Reduced endotoxemia and potentially improved hemodynamic stability and organ function. Demonstrated positive effects on septic shock caused by gram-negative infections. Can stabilize inflammatory and hemodynamic responses in transplant recipients. Some studies suggest improved survival in specific patient populations	Limited evidence in large, well-designed trials. No consistent reduction in mortality across all trials. Uncertainty around optimal timing for initiation of therapy. High cost and logistical challenges of administering multiple sessions. Not effective for all cases of sepsis, especially when endotoxin levels are low
Cytokine adsorption (Cytosorb®)	Removed a broad spectrum of proinflammatory cytokines, potentially preventing or mitigating cytokine storms and graft rejection. Demonstrated high clearance rates of cytokines (> 90%-95%) in vitro. Can be combined with other therapeutic approaches such as continuous renal replacement therapy. May improve hemodynamics and blood lactate levels in some patients	Clinical evidence is limited and inconsistent, with no definitive survival benefit. Primarily used in case reports or small case series, so generalizability is uncertain. Does not target endotoxins, limiting its effectiveness in endotoxin-driven sepsis. Expensive and may require extended treatment (e.g., 6 h per day). No specific cutoff value of interleukin-6 to initiate cytokine adsorption
oXiris® hemofilter (cytokine + endotoxin removal)	Multitasking device capable of removing endotoxins and cytokines and providing renal support. It showed promise in managing septic shock and liver dysfunction post-transplant. Can provide anticoagulation during treatment, reducing the need for additional medications	Clinical outcomes are not consistently improved in large trials. Data from case reports and small studies did not establish clear benefits in post-transplant care. More research is needed to confirm its clinical efficacy and safety in the liver transplant population
Molecular adsorbent recirculation system	Effectively removed toxins and promoted liver function in cases of liver failure. Showed benefits in early allograft dysfunction and primary allograft nonfunction. Decreased serum bilirubin levels, bile acids, serum creatinine, and ammonia levels	Expensive and resource intensive. Long treatment sessions are required (several hours). Evidence for significant survival benefit remains limited. Not universally available, limiting its application in all settings
Therapeutic plasma exchange	Can remove circulating immune complexes, toxins, and cytokines, potentially improving graft survival and reducing inflammation. Established role in managing autoimmune liver diseases, which could benefit patients after transplant. Some studies suggest improvement in post-transplant liver function	Does not directly address the underlying cause of inflammation; only removes circulating mediators. Limited evidence for efficacy in the specific post-transplant setting. Requires multiple sessions, potentially leading to increased costs and complications like transfusion-related acute lung injury, transfusion-associated circulatory overload, etc.
Extracorporeal membrane oxygenation	Provides respiratory and circulatory support in patients with hepatopulmonary syndrome, portopulmonary syndrome, and/or respiratory failure. Can improve oxygenation and hemodynamics in critical patients, reducing the need for mechanical ventilation. Acts as a bridge to recovery	High risk of complications, including bleeding, infection, and organ dysfunction. High resource utilization and intensive monitoring required. Does not address liver dysfunction directly; only provides supportive care