Growing Evidence for Survival Benefit with Plasma Exchange to Treat Liver Failure

Abstract

Plasma exchange (PLEX) to treat liver failure patients is gaining increasing momentum in recent years. Most reports have used PLEX to treat patients with acute liver failure (ALF) or acute on chronic liver failure (ACLF). Etiology of liver disease has an important bearing on the prognosis of the illness in these patients. The accruing data suggest survival benefit with PLEX compared with standard medical treatment to treat ALF and ACLF patients, in randomised controlled trials done world-over. The American College of Apheresis now recommends high-volume PLEX as first-line treatment for ALF patients. Most matched cohort studies done from India which recruited patients with a specific etiology of ALF or ACLF report survival benefit with PLEX compared to standard medical treatment. The survival benefit with PLEX appears more pronounced in ALF patients rather than in ACLF patients. Systematic analysis of the efficacy of PLEX to treat ALF and ACLF patients is needed. There is also a need to identify dynamic predictive scores to assess which patients with ALF or ACLF will respond to PLEX.

Graphical abstract

Plasma exchange (PLEX) to treat liver failure involves two steps—removal of plasma from a patient with liver failure and replacing this with equal volume of fluid; in view of the coagulopathy seen in liver failure patients, the preferred fluid for replacement is fresh frozen plasma. In contrast to hemodialysis, hemadsorption filter or albumin dialysis which remove specific substances (for example: albumin dialysis removes albumin-bound substances), plasma exchange involves the removal of plasma (with all the substances contained in it) and replaces this with healthy donor fresh frozen plasma.

Therapeutic plasma exchange is referred to by many abbreviations in literature like PLEX, TPE or PEX. While PLEX has been used to treat neurological disorders (like Guillain Barre syndrome), hematological disorders (like thrombotic thrombocytopenic purpura) and renal disorders (like hemolytic uremic syndrome) for decades, the use of PLEX to treat liver failure is relatively recent.

Which types of liver failure patients were selected for PLEX treatment in published reports?

Liver failure can be categorised as five phenotypes: hyperacute liver failure, acute liver failure (ALF), subacute liver failure, chronic liver failure (cirrhosis) and acute on chronic liver failure (ACLF).¹ ALF can include hyperacute liver failure and subacute liver failure as per some definitions. As ALF and ACLF patients have high short-term mortality, urgent liver transplantation is needed as a life-saving treatment. PLEX has been recently reported to be beneficial to treat ALF and ACLF patients.

Urgent need for access to life-saving therapies to treat ALF and ACLF patients in India

Viral hepatitis and idiosyncratic drug reaction are major causes of ALF in different parts of India. In southern, eastern and central India, rodenticide ingestion to commit suicide is an important cause of ALF. The true burden of ALF patients in any state of India or in the whole country is not known at present.

Of ∼1000 liver transplants done each year in India, 5–7.5% (ie. 50–75 patients) are performed for ALF patients.¹ It was estimated that rodenticide ingestion (phosphorus poisoning) causing ALF in one state in India (Tamil Nadu) alone would have resulted in the death of 554 patients in the year 2019.²

There is an urgent need to provide liver transplantation as well as non-transplant treatments to save the lives of ALF and ACLF patients.

Different doses of plasma volume exchanged in PLEX to treat liver failure

An average adult has about 5 L of blood (about 2.5 L is plasma). High, standard and low volume PLEX refer to 10 L (4 times the plasma volume), 2.5–5 L (1–2 times the plasma volume) or 1.2 L of plasma (0.5 times the plasma volume) exchanged per session of PLEX.

Different doses of PLEX have been used to treat liver failure patients with high^3,4, standard⁵ or low volume PLEX,⁶ to treat ALF and standard7, 8, 9, 10, 11 or low volume PLEX¹² to treat ACLF. Higher volume of PLEX, adds relatively little to the clearance (and efficiency), but increases the risk of transfusion-related complications like volume overload.¹³

The volume of plasma exchanged in the landmark trial by Larsen et al.³ was decided arbitrarily.¹⁴ It was hypothesised that the failing liver was not able to remove toxic substances which then accumulated in plasma and contributed to the development of multi-organ failure in ALF patients. High-volume PLEX aimed to replace this toxic plasma with healthy donor fresh frozen plasma. Preliminary data suggest that low-volume PLEX may also be beneficial to treat ALF patients.⁶ Apart from strain on blood bank resources (large volumes of fresh frozen plasma needed), risk of transfusion associated acute lung complications, risk of blood borne virus infection, and so on make the use of low-volume PLEX attractive compared to high-volume PLEX.¹⁴

Centrifugal and membrane techniques of PLEX to treat liver failure

Centrifugal or membrane techniques can be used for PLEX. In centrifugal PLEX, the cells in the blood are separated from plasma by molecular-density-based separation during centrifugation (this is the same process used to fractionate blood components in blood banks). In contrast, membrane PLEX uses a filter to separate plasma from cells in the blood (ie. separation based on molecular size). Centrifugal PLEX provides higher plasma removal efficacy than membrane PLEX.¹⁵

Some differences in the two PLEX techniques of relevance in treating liver failure patients are worth noting. Higher rates of blood flow needed for membrane PLEX may result in hemodynamic fluxes which may further adversely affect the perfusion in the compromised hepatic microcirculation. In contrast, the low flow rates with centrifugal PLEX make hemodynamic alterations less likely in liver failure patients. Also, in membrane PLEX, when blood is filtered across a membrane, the already activated blood cells may be further activated, worsening the inflammatory milieu in the liver.

Most studies report the use of centrifugal PLEX to treat liver failure. Preliminary data suggest centrifugal PLEX is superior to membrane PLEX to treat patients with very severe alcoholic hepatitis/alcohol-related ACLF.¹⁶

Does PLEX improve survival in patients with ALF/ACLF ?

Randomised controlled trials (RCTs) provide the highest quality of evidence on whether a therapeutic intervention improves survival. However, the evident benefits as well as challenges in carrying out these trials in patients with ALF have meant that liver transplantation is accepted as standard treatment, without RCTs.¹⁷

Another approach is to compare survival in matched cohorts of ALF or ACLF patients treated with PLEX or with standard medical treatment (SMT). Propensity score matching can be done to better match patients in the two treatment arms.

Etiology impacts prognosis in ALF and ACLF patients

Etiology has a significant bearing on prognosis in ALF patients.¹⁷ The King's college hospital criteria identified ALF due to idiosyncratic drug reaction or non-A non-B viral hepatitis had worse prognosis in non-paracetamol overdose induced ALF.¹⁸ In a multi-centre study, survival rates with native liver were highest in ALF due to paracetamol overdose (57%) or hepatitis A (40%) and least due to Wilson's disease (0% survival).¹⁹

The nature of the acute liver insult influences the severity and outcome of ACLF. Alcohol or cryptogenic ACLF patients had more severe liver disease, higher incidence of organ failures and higher risk of death than hepatitis B- or E-related ACLF.²⁰ In alcohol-related ACLF, metabolic risk factors may impact disease severity and short-term mortality.²¹ Herbal drug-induced liver injury adversely influences survival in alcoholic hepatitis.²²

There are now a plethora of reports on use of PLEX to treat patients with liver failure.

A recent meta-analysis studied RCTs on use of PLEX to treat ALF and ACLF. In this meta-analysis, 4 RCTs (ALF: 2 studies, ACLF: 2 studies) with 736 patients (356 patients had PLEX, 380 patients had SMT) were included. Of ALF patients, 75% was caused by acute viral hepatitis; 57% had advanced encephalopathy; 81% were on mechanical ventilation and 50% on inotropes. This meta-analysis reported better survival at 3 months in patients treated with PLEX than SMT [odds ratio 2.41; 95% confidence interval (CI, 1.28,4.56), P = 0.007].²³

A review of reports comparing PLEX with SMT to treat patients with ALF or ACLF is given in Table 1 (we included RCTs from world-over and matched cohort studies from India). The compiled data in RCTs suggest survival benefit with PLEX to treat ALF or ACLF. Most matched cohort studies of patients of a single etiology (like ALF caused by Wilson's disease⁴ or by rodenticide ingestion [phosphorus poison]⁶ or ACLF caused by alcohol¹²) also show survival benefit with PLEX. The survival benefit with PLEX seems to be more pronounced in ALF than in ACLF patients.

Table 1.

Randomised Controlled Trials (RCTs) Conducted Worldover and Matched Cohort Studies (Cohort) From India on Efficacy of Plasma Exchange (PLEX) to Treat Patients With Acute Liver Failure (ALF) or Acute on Chronic Liver Failure (ACLF). Survival Shown is at 21 days or at End of Hospital Stay in ALF Patients and at 3 months in ACLF Patients.

	Survival with PLEX vs. standard medical treatment to treat liver failure
	Significantly worse	Trend to worsen	Equivocal	Trend to improve	Significantly better
ALF					RCT–13 RCT–25 Cohort–14 Cohort–26
ACLF			Cohort–310 Cohort–411	Cohort–29	RCT–17 RCT–28 Cohort–112

PLEX, Plasma exchange; ALF, Acute liver failure; ACLF, acute on chronic liver failure.

RCTs of PLEX to treat ALF (Table 2)

Table 2.

Randomised Controlled Trials Comparing Survival with Plasma Exchange (PLEX) or Standard Medical Treatment (SMT) to Treat Patients with Acute Liver Failure or Acute on Chronic Liver Failure.

Etiology	Dose of PLEX	Survival without liver transplantation		Increase in survival with PLEX than with SMT	Reference
Acute liver failure
Most common: paracetamol overdose	High volume	In - hospital survival		10.9%	Larsen et al.3
		PLEX (n = 92)	SMT (n = 90)
		58.7%	47.8% (P = 0.008)
Most common: viral hepatitis	Standard volume	21 day survival		30%	Maiwall et al.5
		PLEX (n = 20)	SMT (n = 20)
		75%	45% (P = 0.04)
Acute on chronic liver failure (survival at 3 months)
Hepatitis B	Standard volume	PLEX (n = 104)	SMT (n = 130)	13%	Qin et al.7
		60%	47% (P < 0.05)
Hepatitis B	Standard volume	PLEX (n = 140)	SMT (n = 140)		Yu et al.8
		MELD 30 - 40
		51%	14% (P < 0.01)	37%
		MELD >40
		8%	2% (P > 0.05)	6%

PLEX, Plasma exchange; SMT, standard medical treatment.

ALF: RCT -1

This landmark RCT (Larsen et al.³) showed ∼ 10% increase in survival in ALF patients who had high volume PLEX compared to SMT, among those who did not have urgent liver transplantation. The patients studied had multiple etiologies for ALF; paracetamol overdose was the predominant cause. Two points can be highlighted in this study.

First, this study provides high quality evidence that compared to SMT; PLEX improves survival in ALF patients. Based on the results of this study, American Society for Apheresis now recommends high volume PLEX as a stand-alone treatment for ALF.²⁵

Second, of 183 ALF patients in this study (conducted in three developed countries from 1998 to 2010), only 90 patients (49%) were listed for urgent liver transplantation and only 56 patients (62%) of those listed subsequently had liver transplantation. Also, 64 patients who met poor prognostic criteria and thus merited liver transplantation, were not listed for transplantation due to contraindications (like severe psychiatric or medical co-morbidities). Thus, 127/183 study patients (69%) did not undergo urgent liver transplantation. This highlights the need for non-transplant treatments for ALF.

ALF: RCT -2

In this study, 40 consecutive ALF patients with cerebral edema (noted on computed tomography brain) were randomised to standard volume PLEX or SMT.⁵ The predominant cause of ALF was hepatotrophic viruses (78%) (hepatitis E: 14 patients, hepatitis A: 12 patients, hepatitis B: 5 patients).

Better transplant-free survival at 21 days was noted in PLEX-treated patients (75% survived) than SMT patients (45% survived) (P = 0.04, hazard ratio [HR] 0.30, 95% CI 0.01–0.88). PLEX independently predicted survival (HR, 0.33; 95% CI, 0.11–0.98; P = 0.045) after adjusting for King's College hospital criteria for urgent liver transplantation.

PLEX treatment resulted in greater amelioration of Systemic Inflammatory Response Syndrome (SIRS) and reduction in Sequential Organ Failure Assessment (SOFA) scores than SMT. The investigators presented data suggesting that the improved survival with PLEX was possibly caused by reduction in cytokine storm and ammonia levels.

Matched cohort studies of PLEX to treat ALF from India (Table 3)

Table 3.

Matched Cohort Studies From India Comparing Survival with Plasma Exchange (PLEX) to Standard Medical Treatment (SMT) to Treat Patients with Acute Liver Failure or Acute on Chronic Liver Failure.

Etiology	Dose of PLEX	Survival without liver transplantation		Reference
Acute liver failure
Wilson's disease	High volume	90 day survival		Pawaria et al.4 ILBS Delhi
		PLEX (n = 19)	SMT (n = 18)
		47%	17% (P = 0.049)
Rodenticide (phosphorus) poison in children	Low volume	30 day survivala		Thomas et al.6 CMC Vellore
		PLEX (n = 10)	SMT (n = 8)
		60%	0%
Acute on chronic liver failure (survival at 3 months)
Alcohol	Low volume	PLEX (n = 21)	SMT (n = 29)	Kumar et al.12 CMC Vellore
		38%	14%
Most common: alcohol	Standard volume	PLEX (n = 38)	SMT (n = 38)	Swaroop et al.10 AIIMS Delhi
		63%	47% (P = 0.17)
Alcoholb	Standard volume	PLEX (n = 14)	SMT (n = 14)	Ramakrishnan et al.11 PGI Chandigarh
		64%	50% (P = 0.22)

PLEX, Plasma exchange; SMT, standard medical treatment.

^aSurvival in patients who met Kochi listing criteria for liver transplantation in rodenticidal hepatotoxicity patients was compared with historical cohort.²⁴

^bOnly APASL ACLF grade 2 patients recruited in this study.

ALF: Cohort Study-1

Wilson's Disease Causing ALF

In this study of 37 children with ALF caused by Wilson's disease (median age 9 years, 62% males), 19 children had high volume PLEX and 18 children had SMT.⁴

Liver transplant-free survival at 90 days was better with PLEX (9/19 patients survived (47.3%)) than SMT (3/18 patients survived (16.6%)); (OR 2.84, 95% CI 0.91–8.8, P = 0.049). In Kaplan–Meier analysis, PLEX-treated patients had higher cumulative survival: survival of 38 (12–63) days, median (interquartile range) compared to 14 (5–22) days with SMT. (P = 0.03).

The investigators concluded that in children with ALF caused by Wilson's disease, PLEX may be used both as a bridging therapy to liver transplantation and as a stand-alone treatment, enabling survival with native liver without liver transplantation in some patients.

ALF: Cohort Study -2

Rodenticide Ingestion (Phosphorus Poison)-Induced Hepatotoxicity⁶

Factors predicting poor prognosis and the criteria to list for liver transplantation derived specifically for rodenticidal hepatotoxicity patients (Model for End-stage Liver Disease (MELD) score ≥36 or INR ≥6 at baseline and onset of encephalopathy) were described from Kochi in 2015.²⁴ In this study (which included children and adults), none of the patients who met these criteria and were treated with SMT survived.

A recent study reported improved survival with low-volume PLEX to treat rodenticidal hepatotoxicity in children.⁶ Of 17 children with rodenticidal hepatotoxicity who had PLEX (6 had severe acute liver injury and 11 had ALF), one-month survival was 87.5%. Ten children fulfilled the Kochi listing criteria for urgent liver transplantation, of these, two children had hypotension needing inotropes and were ineligible for PLEX. Of the remaining eight children who met Kochi listing criteria and were eligible for PLEX, six children (75%) were alive one month after PLEX.

It is possible that the improved survival in PLEX-treated patients may have been due to other improvements in the care of critically ill patients as the two above studies were separated almost by a decade.

RCTs of PLEX to treat ACLF (Table 2)

ACLF: RCT -1

PLEX to Treat Hepatitis B-Related ACLF⁷

In this study conducted from 2003 to 2007, 234 patients with hepatitis B-related ACLF who were not eligible for liver transplantation were randomised to PLEX (n = 104) or SMT (n = 130).⁷ ACLF was diagnosed as per Chinese Society of Hepatology definition. Nucleoside analogues were given to patients with HBV DNA level >1000 copies/mL. Study patients were followed up for ≥5 years or until death.

In the PLEX arm of the study, all patients with coagulopathy were treated with PLEX. For patients with encephalopathy, hemoperfusion or continuous hemodiafiltration was suggested, in addition to PLEX. For patients with hepatorenal syndrome or electrolyte imbalance, continuous hemodiafiltration was suggested in addition to PLEX. The 104 PLEX-treated patients had 2 (1–8; mean, range), PLEX sessions. In addition to PLEX, 11 patients had hemoperfusion and 4 patients had continuous hemodiafiltration.

Better 90-day survival was noted with PLEX (60%) than SMT (47%), (P < 0.05). Survival at 5 years was also better with PLEX (43%) compared to SMT (31%), (P < 0.05). Median survival in PLEX-treated patients was 879 days and in SMT-treated patients was 649 days.

PLEX treatment predicted better survival on multivariate analysis.

ACLF: RCT -2

PLEX to Treat Hepatitis B-Related ACLF⁸

In this study, 280 patients with hepatitis B-related ACLF were randomised to PLEX or SMT.⁸ ACLF was diagnosed as per the Chinese Medical Association definition. All patients were treated with Lamivudine. Over the 3 months of follow-up, 222 patients died.

In patients with MELD score from 30 to 40, survival at 3 months was better with PLEX (49.4% died) than SMT (86.1% died) P < 0.001. In patients with MELD score >40, there was no difference in survival at 3 months in the two treatment arms (PLEX treated: 91.5% died compared to SMT: 98.4% died, P > 0.05).

The authors reported a significant decrease in 3 month mortality with PLEX treatment in hepatitis B-related ACLF patients who had MELD scores between 30 and 40. In these patients, low baseline HBV-DNA level and rapid fall in HBV-DNA level were good predictors of survival with PLEX and lamivudine.

Matched cohort studies of PLEX to treat ACLF from India (Table 3)

ACLF: Cohort Study-1 (Table 2, Table 3)

PLEX to Treat Alcohol-Related ACLF¹²

In this study of 50 alcohol-related ACLF patients, conducted from 2017 to 2019, 21 patients who had low-volume PLEX were compared to 29 patients treated with SMT.¹² ACLF was diagnosed as per Asia Pacific Association for the Study of the Liver (APASL) definition. All 50 patients had discriminant function [DF] score >32 (ie. they had severe alcoholic hepatitis).

The primary study outcome was one-year survival.

PLEX-treated patients underwent 3 (1–7) PLEX sessions with 1.5 (1.4–1.6) litres of plasma exchanged per session. They also received Prednisolone tablets 20 mg daily, tapered over 1 month.

Survival estimates (standard error) with PLEX at day 30 was 66% (0.11), day 90 was 35.6% (0.11), 6 months and one year was 20.3% (0.09) and with SMT at day 30 was 16.6% (0.08), day 90, 6 months and one year was 11% (0.07). In Kaplan–Meier analysis, PLEX resulted in better survival over 1 year than SMT (P = 0.03). HR (adjusted for baseline MELD score) for survival favoured low-volume PLEX and low-dose steroid (0.44 (0.22–0.86)) compared to SMT.

Ten patients (48%) in PLEX group had renal dysfunction; this normalized following PLEX in 6 patients (60%).

ACLF: Cohort Study-2

PLEX to Treat ACLF (Predominantly Alcohol-Related)⁹

In this study, a multinational cohort of 1866 ACLF patients from the APASL ACLF Research Consortium (AARC) database was analysed to compare survival in patients treated with PLEX to propensity score matched patients treated with SMT.⁹ ACLF was diagnosed as per APASL definition.

In the propensity score matched cohorts, 94 patients had PLEX and 89 patients had SMT. At 3 months, there was a trend to better survival with PLEX than SMT (P = 0.08).

PLEX improved systemic inflammation in these patients. Competing risk analysis found that PLEX therapy resulted in less deaths due to liver failure at 90 days.

ACLF: Cohort Study-3

PLEX to Treat ACLF (Predominantly Alcohol-Related)¹⁰

In this study, survival was compared in ACLF patients treated with PLEX or with SMT—both groups of patients were propensity score matched for sex, ACLF grade, CLIF-C ACLF scores and hepatic encephalopathy.¹⁰ ACLF was diagnosed as per European Association for the Study of the Liver (EASL) definition.

Primary study outcomes were mortality at 30 days and at 90 days.

Of 1151 patients, ACLF n= 864 patients [75%] and AD [without organ failure] n = 287 patients [25%] were included. Of the patients with ACLF (n = 864), grade 1, 2 and 3 ACLF was present in 167 (19.3%), 325 (37.6%) and 372 (43.0%) patients, respectively. In the propensity-matched analysis (38 patients treated with PLEX and 38 patients treated with SMT were included), 30-day mortality was lower with PLEX (21%) than SMT (50%), P = 0.01. However, at 90-days, mortality was not significantly different with PLEX (36.8%) or with SMT (52.6%, P = 0.17); HR, 0.82 (0.44–1.52), P = 0.55.

The 30 day survival with PLEX treatment was significantly better in ACLF patients than in patients with acute decompensation of cirrhosis.

The authors concluded that PLEX improved short-term survival in ACLF patients, but did not significantly impact long-term survival.

ACLF: Cohort Study-4

PLEX to Treat Grade 2 Alcohol-Related ACLF¹¹

In this study, alcohol-related ACLF grade 2 (ie. AARC score 8–10) patients were enrolled, if they did not respond to SMT for 1 week (ie. failure to improve AARC score by ≥ 2 points) and did not have an immediate option of liver transplantation.¹¹ ACLF was diagnosed and ACLF grade was categorised as per APASL definition. Of the 28 patients enrolled into the study, 14 patients had PLEX and 14 patients had SMT. Five PLEX sessions were planned in the PLEX-treated patients.

The mortality at 90 days was higher with SMT (50%) than with PLEX (36%); however, this difference was not statistically significant (P = 0.22). PLEX treatment resulted in reduction in liver disease severity scores like AARC, DF score and MELD scores.

In the 9 patients who had the total 5 PLEX sessions as planned, the 30-day mortality was nil compared to 21% with SMT (P = 0.07) and the 90-day mortality was 11% compared to 50% with SMT (P = 0.03).

Indications for PLEX to Treat Liver Failure Patients

The published reports suggest benefit with PLEX to treat ALF, severe acute liver injury and ACLF of any etiology.

The American College for Apheresis recommends high-volume PLEX as first choice of therapy (either as stand-alone treatment or along with other modes of treatment) for ALF.²⁵

The Tamil Nadu chapter of Indian Society of Gastroenterology (TN-ISG)²⁶ as well as National Health Mission – Tamil Nadu Accident and Emergency Care Initiative (NHM–TAEI)²⁷ guidelines specify indications and contra-indications for PLEX to treat rodenticide (phosphorus) ingestion induced hepatotoxicity. These guidelines recommend that any patient with rodenticidal hepatotoxicity with severe acute liver injury or ALF (with INR ≥4, serial increase in INR or hepatic encephalopathy) be treated with PLEX. Low-volume PLEX is the preferred dose of PLEX advocated by the TN-ISG as well as by NHM–TAEI.

Thus, many ALF patients who meet the King's college criteria for urgent liver transplantation¹⁸ are now treated by PLEX, either as a bridge to liver transplantation or as stand-alone treatment.

PLEX provides survival benefit in ALF and ACLF—these syndromes are characterised by overactive innate immune response.28, 29, 30 It is important for the clinician to identify the patients with liver failure who have overactive innate immune response. Serum ferritin and plasma von Willebrand factor (VWF) levels are useful markers of reticuloendothelial activation which helps guide treatment in these patients.³¹

“Traffic jam in liver sinusoids” hypothesis to explain how PLEX helps ALF and ACLF patients to recover

How does the failing liver in ALF and ACLF recover with PLEX ? Early leads suggest that PLEX ameliorates the overactive innate immune and inflammatory response and reverses failure/dysfunction of liver and other vital organs in these patients. The clogging of acute inflammatory cells and debris can lead to sludging in the liver sinusoids (like a “traffic jam”) and can cause a functional acute sinusoidal obstruction, which in turn reduces perfusion in liver microcirculation and causes the liver to fail.32, 33, 34

In contrast to dialysis, which did not improve survival in ALF patients, PLEX improved survival in ALF patients. This suggests that the survival benefit with PLEX maybe by removal of substances in the plasma which are too big to be dialysed (ie. > 60 -kDa in size). VWF is a candidate macromolecule in this regard. VWF molecules are the largest sized plasma protein in health and raised plasma VWF levels are a MELD-independent predictor of short-term mortality in ALF and ACLF patients. PLEX may clear inflammatory macromolecules like VWF and help restore perfusion in hepatic microcirculation, thus restoring liver function in ALF/ACLF patients.34, 35, 36, 37

Similarly, other macromolecules may clog the filter in other vital organs in these patients. A variety of macromolecules are upregulated in ACLF patients who develop acute kidney injury compared to those who do not develop acute kidney injury.³⁸ It is possible that the reversal of renal failure with PLEX to treat alcohol related ACLF patients¹² may be partly due to the removal of such inflammatory macromolecules.

Contra-indications to PLEX to treat liver failure patients

Few relative contra-indications to the use of PLEX to treat ALF or ACLF patients have been proposed. These contra-indications are hemodynamic instability, active sepsis and recent gastrointestinal (GI) bleed.

Hemodynamic instability was considered as contra-indications for the use of PLEX in some reports.^6,12 The concern raised is that removal of fluid/plasma from a patient who is already hemodynamically unstable may worsen this and adversely affect patient outcomes. On the other hand, studies have reported (transient) improvement of hemodynamic indices after PLEX to treat ALF patients.³⁹

Active sepsis is considered a contra-indication to PLEX. It is possible that dampening the overactive immune response by PLEX may worsen sepsis. The golden window of sterile inflammation in ACLF and in ALF patients maybe the ideal setting for initiating PLEX.⁴⁰

Recent GI bleed has been considered a contra-indication to the use of PLEX to treat liver failure in some reports.¹²

Low-volume PLEX and low-dose steroid to treat liver failure: the Vellore protocol

The technical details of our protocol using low-volume PLEX and low-dose steroid to treat patients with liver failure have been published.³² We have summarised the key points of this protocol in Table 4. The use of low dose steroid and, if possible, avoiding the use of sedative drugs are advocated in this protocol.

Table 4.

Low-volume Plasma Exchange (PLEX) and Low-dose Steroid to Treat Liver Failure: Do's and Don'ts From the Vellore Protocol.³²

	Do's
	1. Admit to closely monitored setting under dedicated multi-disciplinary team. 2. Ultrasound guided PLEX port insertion into femoral vein preferred. 3. Oral 10 mg prednisolone once daily started a day prior to initiating PLEX. 4. Centrifugal PLEX preferred, done daily, for 3 days (50% plasma volume exchanged per session, flow rate: 25–40 ml/min). 1:1 volume replaced with fresh frozen plasma (FFP). Calcium (oral/IV) given during PLEX to avoid citrate toxicity. Further sessions decided by treating physician as per clinical assessment. 5. Take surveillance blood culture. Give prophylactic antibiotics as follows: if no/suspected sepsis: IV Cefaperazone-Sulbactam; if suspected sepsis with organ dysfunction: IV Meropenem. 6. Enteral nutrition preferred. 7. Follow up: a) Steroid given, balancing its anti-inflammatory action and risk of sepsis, after periodic clinical assessment. b) OPD visit - once weekly for 4 weeks, then once a month for 3 months, then as needed.
	Don'ts
	1. Better to avoid sedatives (may precipitate/worsen encephalopathy, need mechanical ventilation). If needed, use sedative at lower than usual dose at lesser frequency. 2. Avoid prophylactic blood product transfusion. If platelets are being transfused, transfuse FFP prior to platelets (to supplement ADAMTS13). 3. Avoid potentially hepatotoxic drugs 4. Defer invasive tests/tests needing shifting patient which can wait till patient stabilises.

PLEX, Plasma exchange

Rapidly upscaling access to PLEX to treat rodenticidal hepatotoxicity in Tamil Nadu: interventions by government of Tamil Nadu

A study conducted by TN-ISG in 6 districts of Tamil Nadu in 2019 documented rodenticide ingestion as the commonest cause of acute toxic hepatitis; in contrast, paracetamol overdose is the commonest cause in Western countries. The ratio of patients with hepatotoxicity due to paracetamol overdose to rodenticidal hepatotoxicity was 1 : 75 in this TN-ISG study (Table 5).²

Table 5.

Improving Access to Urgent Life-saving Treatments for ALF Patients—The Example of Rodenticidal Hepatotoxicity in Tamil Nadu.

Rodenticide (phosphorus) poison ingestion		Tamil Nadu
		In 6 districts: TN-ISG study2	in entire state
			Estimated2	Actual
No. of patients		Jan–Jun 2019	Jan–Dec 2019	Jan–Dec 2022
	Took poison	659
	Developed liver damage	450	1584 (95% CI: 265–6119)
	Died/poor outcome	159 (35%)	554
	Treated by PLEX	54		652a
	Had urgent liver transplantation	1		2b

TN-ISG: Tamil Nadu chapter of Indian Society of Gastroenterology, PLEX: plasma exchange.

^aUnder National Health Mission (Tamil Nadu Accident and Emergency Care Initiative).

^bData from Transplantation Authority of Tamil Nadu (TRANSTAN).⁴²

In this study, 450 rodenticidal hepatotoxicity patients (25, 21–34 years old) were treated with SMT (396 patients), PLEX (54 patients) and liver transplantation (1 patient); 159 patients (35%) had poor outcome (131 died, 28 discharged in terminal state).

As demonstrated in this study, urgent liver transplantation is not accessible to most patients in Tamil Nadu at present. This stark reality is probably true for the entire Indian population. The public sector hospitals see a huge burden of rodenticide poisoning patients.⁴¹ During the 180 days of the TN-ISG study, two participating government medical college hospitals cared for >180 rodenticide poison patients (ie. ≥ 1 patient per day).²

In response to the above study findings, TN-ISG formulated guidelines to manage rodenticide poison focussing on non-transplant treatments.²⁶ It was proposed that each district of Tamil Nadu should provide a “hub and spoke” model of care—patients receive first aid and decontamination of poison (like gastric lavage) at primary health centres; they are referred to government taluk hospitals for observation for 5 days; those who develop liver damage are referred to government medical colleges where specialised care including PLEX are provided. Probably for the first time in the world, indications and contra-indications for PLEX to treat ALF or severe acute liver injury were proposed.

The Tamil Nadu government initiated a series of interventions to address this public health problem. Steps to restrict the sale of rodenticide were initiated. NHM–TAEI brought out guidelines on managing rodenticide and other poisons,²⁷ trained teams in government hospitals who manage these patients and set up facilities for specialised care including PLEX in government medical college hospitals (Table 5).

PLEX to treat other etiologies of ALF and ACLF

Benefits with PLEX are now reported to treat ALF due to varied causes like idiosyncratic drug reaction^33,43, pregnancy-related liver diseases44, 45, 46 (acute fatty liver of pregnancy, HELLP syndrome), non-hepatotrophic viruses (dengue,³² parvovirus B19⁴⁷), poisons/toxins^48,49, and early allograft dysfunction⁵⁰ and thrombotic microangiopathy⁵¹ after liver transplantation. PLEX may also benefit ACLF patients with viral hepatitis complicating non-alcoholic fatty liver disease⁵² or Wilson's disease.⁵³

High-dose steroid versus low-dose steroid and low-volume PLEX to treat alcohol-related ACLF

Alcohol-related ACLF patients have high short-term mortality and are poor candidates for steroid therapy. Many patients with severe alcoholic hepatitis also meet criteria for ACLF. While patients with severe alcoholic hepatitis may be treated with Prednisolone 40 mg once daily, patients with alcohol-related ACLF tolerate steroid poorly. It is uncertain if patients with severe alcoholic hepatitis and ACLF should be treated with steroid (it is not clear if steroids are beneficial or harmful in them).⁵⁴ Renal failure is considered a contra-indication for the use of steroid to treat severe alcoholic hepatitis patients.⁵⁵

Preliminary data suggest that low volume PLEX and low-dose steroid (Prednisolone 10–20 mg once daily) improves 1 year survival compared to SMT in patients with alcohol-related ACLF patients (who also had severe alcoholic hepatitis). Raised serum creatinine is not considered contra-indication to PLEX treatment in these patients; in fact, serum creatinine normalised in 6/10 patients treated with low-volume PLEX and low-dose steroid.¹²

When is it futile to treat liver failure patients with PLEX?

In ALF patients with clinical features of brain herniation (like unequal or fixed dilated pupils), it is probably futile to perform PLEX. Similarly, in ALF or ACLF patients with >2 extra-hepatic organs failing, PLEX may be futile.

Complications of PLEX to treat liver failure

PLEX to treat patients with ALF and ACLF appears generally well tolerated. The complications of PLEX reported include local complications like port-related infection, thrombophlebitis, bleed, inadvertent arterial puncture; transfusion-related acute lung complications like transfusion associated circulatory overload (TACO), transfusion-related acute lung injury (TRALI); sepsis and citrate (anti-coagulant) toxicity.

TACO and TRALI are potentially fatal complications. It is likely that TACO and TRALI may be seen more often in ALF and ACLF who are already mechanically ventilated. In contrast, in ACLF patients who were not on mechanical ventilation or oxygen support treated with low-volume PLEX, TACO and TRALI are uncommon. This points to a safe window to help guide patient selection for this treatment.⁵⁶

Blood taken out through an extra-corporeal circuit during PLEX can clot. The choice of anti-coagulation in the coagulopathic ALF or ACLF patient needs attention.⁵⁷ Regional anti-coagulation (ie. anti-coagulation within the extra-corporeal circuit) is the ideal method. Most centres prefer not to use heparin in liver failure patients. Citrate (a calcium chelator) is used by most centres for regional anti-coagulation. As citrate is metabolised by the liver, there is a concern of citrate toxicity in patients with a failing liver. Citrate toxicity manifests as symptoms of hypocalcemia (tingling, tetany, etc).

Prophylactic calcium supplementation during PLEX with calcium monitoring helps prevent hypocalcemia. With low volumes of plasma exchanged in low-volume PLEX (thus the amount of citrate used is less) and with low rates of flow of the processed blood (with centrifugal technique of PLEX), citrate toxicity appears uncommon in liver failure patients treated with low-volume PLEX.

PLEX to treat special populations

PLEX has been used to treat ALF in children^4,6,51 and pregnant women.46, 45, 44

Limitations of current data on PLEX to treat ALF and ACLF patients

The published reports on PLEX to treat ALF and ACLF have used different doses of PLEX. ACLF is defined by different definitions in these reports. Most studies have enlisted patients with ALF or ACLF caused by multiple etiologies. These are some of the challenges in interpreting survival outcomes of PLEX to treat ALF and ACLF.

The reports of the use of PLEX to treat liver failure are rapidly increasing; however, the data on outcomes of PLEX for specific etiologies of ALF or ACLF are still scarce. The possibility of publication bias also needs to be kept in mind, especially in the case of cohort studies. RCTs with large sample sizes are needed. It is possible that the improvement in survival with PLEX may be also due to other interventions given as part of PLEX protocol like low-dose steroid and avoidance of sedation (as discussed below).

Most sedative drugs are metabolised in the liver. In patients with acute liver dysfunction, administration of sedative may result in prolonged drug half-life, unintended sedative overdose leading to respiratory depression and may precipitate or aggravate encephalopathy.^58,59 The guidelines issued by TN-ISG²⁶ and by NHM–TAEI²⁷ on managing rodenticide poison patients recommend avoiding sedative drugs or restricting its use to select indications, at low doses, when needed. Most doctors working in resource-constrained settings endorse this practice as well.⁶⁰

Dynamic scores needed to better predict survival in ALF and ACLF patients treated with PLEX

The rapidly changing course of illness is more accurately predicted by dynamic rather than static scores in ALF⁶¹ and in ACLF patients.^62,63 Dynamic scores are needed to assess prognosis after PLEX to treat ALF and ACLF. Should all patients listed for urgent liver transplantation be treated with PLEX ? Is PLEX a stand-alone treatment or bridge to liver transplantation? Which clinical or laboratory signals indicate that a patient will survive with native liver after PLEX ? As the window for therapeutic interventions is short, these dynamic parameters need to be done soon after PLEX.

It is recommended that all blood samples for tests be drawn at least 12 h after completion of a PLEX session.²⁵ Post-PLEX dynamic predictive scores need to use these timed indices for the score calculation. A patient who improves with PLEX may not need liver transplantation.

Where does PLEX figure in the order of priority of treatments for liver failure patients?

In contrast to urgent liver transplantation, PLEX is accessible to many more patients in India and world-over. Should PLEX be offered to all patients with ALF or ACLF ? In patients listed for urgent liver transplantation, will PLEX help stabilise the patient or will this only drain more resources as well as interfere with the decision making process in patient management? These important questions need deliberation by experts, in view of the rapidly accruing data on improved survival with PLEX to treat ALF and ACLF (Table 6).

Table 6.

Different Steps in Treating Acute Liver Failure World-over at Present—Where is PLEX to be Offered?

Access to urgent LT	Urgent treatment provided
Country with predominant DDLT (ex: UK, USA, Europe, Australia)	Urgent LT	If contra-indications to LT, consider PLEX
Country with predominant LDLT (ex: Korea, India): if patient has access to LDLT	PLEX	Urgent LDLT
	Urgent LDLT
Country with no LT (ex: Africa, many SAARC nations other than India) or very little access to urgent LT (most of Indian population)	PLEX
	SMT

LT: Liver transplantation, DDLT: deceased donor liver transplantation, LDLT: live donor liver transplantation, SAARC: South Asian Association for Regional Co-operation, PLEX: plasma exchange, SMT: standard medical treatment.

We propose an algorithm for PLEX treatment in liver failure syndromes (Figure 1), wherein patients with less severe disease are treated with PLEX and those with more severe disease are treated by liver transplantation. PLEX can either be a stand-alone treatment or act as a bridge to liver transplantation.

Figure 1.

Proposed algorithm for treatment of ALF and ACLF patients: when to perform plasma exchange (PLEX) ? Patients with less severe liver disease may respond to PLEX treatment, while those with more severe disease will need liver transplantation. Those who do not respond to PLEX may be considered for liver transplantation. Dynamic predictors of response to PLEX are urgently needed. ALI: acute liver injury, ALF: acute liver failure, ACLF: acute on chronic liver failure.

Combining PLEX with other modalities to treat liver failure patients

It is likely that in future, a combination of treatments like PLEX and continuous venovenous hemofiltration (CVVH) to treat ALF¹⁴ or PLEX and sustained low efficiency dialysis will be tailored for specific liver failure patients. These algorithms have to evolve based on clinical studies.

In liver failure patients, while the terms continuous renal replacement therapy and CVVH have been loosely used inter-changeably to refer to the same treatment, we prefer the term CVVH. The indications for CVVH to treat ALF patients are classified into non-renal indications (like hyperammonemia) and renal indications (like hyperkalemia). It is accurate to term this treatment of ALF to lower ammonia levels as CVVH, in the absence of renal dysfunction.

What are the requirements of studies to assess survival in ALF or ACLF?

More RCTs or matched cohort studies in ALF or ACLF patients are needed to compare survival with PLEX with other modes of treatment.

Recruiting patients with a single etiology of ALF or of ACLF for these studies appears to be important. As etiology impacts the prognosis of ALF and ACLF, prognostic scores of liver failure specific to a particular etiology are likely to be more accurate.

It is preferable to report survival at 1 month in ALF patients and at 3 months or 1 year in ACLF patients. The need for dynamic scores to predict survival after PLEX to treat ALF and ACLF is gaining more urgency.⁶⁴ It is important to be able to predict which patient with ALF or ACLF is responding to PLEX or not, especially to decide on listing for urgent liver transplantation.

While more lives are saved with PLEX treatment in ALF and ACLF patients than SMT, it is a sobering fact that significant proportion of patients do not respond to PLEX and still die in the short-term.⁶⁵ This highlights the need for improvements to the PLEX protocol and addition of other adjuvant therapies in these patients.

This article comments on the evidence to support the recently increasing use of PLEX to treat ALF or ACLF patients. The limited data in published literature points to survival benefit with PLEX. There are many deficiencies in the current evidence base which need to be systematically addressed. The rapidly accruing data need to be assimilated to clarify when PLEX is to be initiated to treat ALF and ACLF patients. It is likely that ALF and ACLF patients with less severe disease may be rescued with PLEX and those with more severe disease will need urgent liver transplantation.

Author contribution

AG, UZ, DD, CEE are involved in conceptualizing, drafting and final approval of the manuscript.

Conflicts of interest

The authors have none to declare.

Funding

None.