Systematic Review
Copyright ©The Author(s) 2020.
World J Gastroenterol. Jan 14, 2020; 26(2): 219-245
Published online Jan 14, 2020. doi: 10.3748/wjg.v26.i2.219
Table 1 Participants, interventions, comparisons, outcomes and study design criteria used to define the research question for this systematic review
VariableDescription
PopulationHumans diagnosed with liver failure (ALF/ACLF)
InterventionPlasma exchange with or without other alternative liver support systems; no restrictions on dose, duration and type of plasma exchange
ComparatorRandomized controlled trials/Cohort studies: Standard medical treatment
Case series/case reports: Nil
OutcomeAll-cause mortality, changes in liver biochemistry, and survival in non-transplanted patients
Study designRandomized Controlled Trials, Cohort studies, Case series, Case reports
Research questionDoes plasmapheresis have an effect on all-cause mortality, changes in liver biochemistry, and survival in non-transplanted patients with ALF/ACLF, compared to standard medical treatment?
Table 2 Studies included for study of plasmapheresis in acute liver failure in adults
Ref.Type of study/No. of patients recruitedStudy group(s)Plasma exchange regimeEtiology of liver failureResults/outcome(s) of interest
Larsen et al[3]Open randomized control trial;PE + SMT vs SMTPlasma exchange volume: Volume of plasma exchange was 15% of ideal body weight (representing 8-12 L per day per procedure); patient plasma was removed at a rate of 1-2 L per hour with replacement with fresh frozen plasma in equivalent volumePredominantly paracetamol (59%), followed by unknown etiology, toxic hepatitis, viral hepatitis, acute Budd Chiari syndromeHVP increases transplant free survival after 3 mo, and maximal effect of HVP was achieved in patients who did not undergo emergency transplantation
n = 182 (Plasma exchange + SMT 92, SMT 90)
Overall hospital survival was 58.7% for patients treated with high volume plasma exchange vs 47.8% for control group HR: 0.56 (95%CI: 0.36-0.86), P < 0.01
However, HVP prior to transplantation did not improve survival compared with patients who received SMT alone P = 0.75
HVP procedure was undertaken on three consecutive days but with no fixed time interval between each treatment
Bilirubin, INR, ALT and ammonia concentration decreased significantly during the first 7 d compared to SMT
Mean number of HVP = 2.4 ± 0.8
Plasma exchange with donor plasma
Nakae et al[10]Prospective cohort study; n = 13PE vs PE + CHDFPlasma exchange volume: 3.6 to 4.0 L of plasma was exchanged for the same volume of normal FFP, interval between sessions was 48h or more for both treatment methods7 post surgery, 4 fulminant hepatitis, 4 sepsisTotal bilirubin levels were significantly lower after treatment in both arms: Both P < 0.01 in both groups
No outcomes available on mortality
Of note, decreased increase in citrate in patients with PE CHDF compared to PE alone
Plasma exchange with FFP
Hung et al[5]Retrospective cohort study; n = 62 (Control 32, PE 30)PE vs controlAverage plasma exchange volume: 2916 mL (range, 1875-3750 mL), plasma exchange occurred over 2 h46.7% HBV, 33.3% Drug induced, 6.7% unknown, 33.3% cirrhosis.At the end of the first week (week 1), the level of total bilirubin and grading of hepatic encephalopathy in the PE group were significantly lower than those in the control group. At the end of the second week (week 2), there were no differences in the level of total bilirubin and grading of hepatic encephalopathy between the two groups of patients
Difference in survival rate was not significant 66.7% in PE group vs 59.4% in control group
No significant differences in etiology of liver failure between treatment and control groups
Average of 6 rounds of exchange per patient (range, 2-15 rounds)
Difference in survival days were significant, with 17.63 ± 1.86 in PE group vs 8.69 ± 0.86 in control group. P = 0.01
Plasma exchange with FFP
Li et al[4]Retrospective cohort study; n = 61PE + HP + CVVHDF, PE + CVVHDF and HP + CVVHDFPlasma exchange volume: 2000-3000 mL of fresh per session. Flow rate was 80-120 mL/min, the plasma separation rate was 25-30 mL/min, the replacement time was 2.0-3.0 h3/61 acute viral hepatitis, 17/61 chronic toxic acute liver failure. 41 cases of non-viral induced liver injury: 5 after cardiac surgery, 7 with drug poisoning, 13 cases after pregnancy childbirth, 1 case mushroom poisoning, 10 cases with severe infection, 5 othersTreatment of the 61 patients using the artificial liver support system yielded a survival rate of 62.3% (38/61), and a viral survival rate of 35.0% (7/20); with the non-viral survival rate being 75.6% (31/41) Biochemically PE + HP + CVVHDF and PE + CVVHDF groups saw improvement in total bilirubin, ALT, PT, Albumin and HP+CVVHDF saw improvement in total and ALT (P < 0.05)
In the PE + HP + CVVHDF group: After completion of a single plasma exchange, the HP was carried out. After HP, the CVVHDF is carried out
Total of 171 exchanges were done
Nakamura et al[11]Retrospective case series n = 49; Fulminant hepatitis 15; Severe acute hepatitis 14; Healthy controls 20No comparative armPlasma exchange volume: Approximately 2000–4000 mL of fresh-frozen plasma was substituted during each exchangeNo mention10/15 fulminant hepatitis and all severe acute hepatitis survived
Significant decreases in circulating TNF-a, IL-6, and TGF-ß levels in patients with fulminant hepatitis after a single plasma exchange
Plasma exchange with FFP
Akdogan et al[9]Retrospective case series; n = 39 (fulminant hepatic failure)PE, No comparative armPlasma exchange volume: Total plasma volume approximately 1Predominantly undetermined (41%), paracetamol (28.5%), acute hepatitis B, autoimmune liver disease, vascular tumor, acute hepatitis A (in presence of cirrhosis)Improved biochemically (coagulopathy hyperbilirubinemia, AST, Ammonia, Factor V levels): P < 0.05
Plasma exchange continued on a daily basis till clinical response (subjective by ICU team) or patient expired, or transplanted31% underwent liver transplant, 92% of which survived at 1 year
Overall survival 54% (21/39 patients), 37% (10/27) of non-transplanted patients survived
No need for calcium replacement or magnesium replacement
Plasma exchange with low volume citrate plasma
Kondrup et al[6]Case series; n = 11PE, No comparative armPlasma exchange volume: 20% body weight plasma exchange intended on three consecutive days, obtained a mean 2.6 exchanges and mean volume 16% body weight6 acetaminophen, 2 non-A/B hepatitis, Halothane, disulfurum toxicity and hepatitis B5/11 survivors were all acetaminophen toxicity induced ALF
All had improved bilirubin after treatment
All 4 Grade IV encephalopathy patients all did not survive
Plasma exchange with donor plasma.Those that survived had Grade III encephalopathy or lesser
Those that did not survive had a longer duration of coma before initiation of PE (3.5 vs 1.8 d)
Freeman et al[13]Case seriesPE, No comparative armPlasma exchange volume: 3 L of plasma exchange was performed daily until conscious level improved or patient died Plasma exchange fluid: Equal volume of compatible fresh frozen plasma and plasma protein fraction (PPF) usually in the proportion of 2 units FFP:1 PPF4 acetaminophen, 2 nonA/B hepatitis, 1 hepatitis A, 1 mixed drug overdose, 1 ETOH in 27/9 showed improvement in coma grades, 5 achieved normal mental state, 5 were able to discharge from hospital. Of which 3 were paracetamol induced liver failure, 1 was monoamine oxidase/ tricyclic acid induced, 1 was alcohol. Survival 55%
n = 9
Improved biochemistry, bilirubin, coagulation (P < 0.01) 1 died of retroperitoneal bleeding
Buckner et al[7]Case seriesPE, No comparative armPlasma exchange volume: Initial 10 L/day plasma exchange with FFP or fresh/outdated plasma1 Acute viral hepatitis (pediatric), 2 halothane, 1 hepatitis B viral hepatitis1 died (pediatric)
1 patient took 37 d to awake from coma
n = 4 (1 pediatric)
3/4 of patients survived
Liu et al[31]Case seriesPE, No comparative armPlasma exchange volume: Each treatment lasted for 4-6 h, and the total volume exchanged was approximately 7000 mL (1.5-2x TPV)DILIThe two patients with DILI ALF were treated with PE without need for transplant
n = 2
Biochemically improved after PE (AST ALT Bilirubin)
Plasma exchange fluid: Maximally, 4700 mL of FFP was exchanged in each session, and the rest comprised plasma substitute consisting of 25% human albumin, pentastarch, 0.9% saline, and Ringer's solution. In each session, the plasma substitute was exchanged initially, and FFP was exchanged at the end
Duration of PE was based on clinical improvement, both patients had intermittent PE, total 3 sessions
Bilgir et al[15]Case reportPE, No comparative armPlasma exchange volume: Each session consists of 15 units of FFP, total 4 sessionsL-asparaginase induced ALFPatient recovered from ALF: However, no biopsy done
Plasma exchange fluid: FFP
Aydemir et al[14]Case reportPE, No comparative armPlasma exchange volume: 2500 mL plasma volume removed during each PE sessionPTU induced ALFPatient recovered from ALF: however, no biopsy done
Plasma exchange fluid: Fresh frozen plasma
Riveiro-Barciela et al[28]Letter to editor, case report (Ipilimumab)PE, No comparative armPlasma exchange volume: 1500 mL of 5% albumin plus 4 units of plasma as replacement fluid, carried out every other day for total 5 treatmentsImmunotherapy induced ALFPatient improved. Liver tests within normal values within one month
Plasma exchange fluid: FFP and 5% albumin
Damsgaard et al[8]Case report (ALF in WD)PE, No comparative armPlasma exchange volume: 8-9 L of plasma, total 12 HVPFulminant Wilson’s disease ALFEven though WD ALF score was 16, patient survived without need for OLT
Plasma exchange fluid: Fresh frozen plasma as replacement fluid 1:1
Göpel et al[33]Case report (Letter to editor)PE, No comparative armPlasma exchange treatment was performed for three consecutive daysPeg-asparaginase induced ALFPatient improved. Continuous stabilization of fibrinogen and antithrombin 3, an increase of cholinesterase, and a decrease of bilirubin. Clinical signs and symptoms such as jaundice and ascites did also rapidly improve
No mention of volume or type of exchange fluid
Lin et al[16]Case reportPE, no comparative armPlasma exchange was performed 2 times per week, and 2000 to 2500 mL frozen plasma was used each timeHLHPatient’s condition deteriorated, and he died of multi-organ failure during the 6th week of hospitalization. Autopsy was declined
Chen et al[29]Case reportPE, No comparative armPlasma exchange volume: Estimated two times the plasma volume of the patient was exchanged. At most, 40 units of FFP were exchanged, with the remainder of the infused volume consisting of plasma substitutes. The plasma substitutes consisted of 25% human albumin, pentastarch, normal saline, and Ringer’s solutionHeat strokeOn day 4 after the admission, the patient received high-volume PE (two plasma volumes exchanged). His consciousness was improved a day after PE The patient was discharged on day 16 after admission without sequelae
Holt et al[17]Case reportPE, No comparative armPlasma exchange on post-partum days 3-5. Volume: Average of 3.2 L (1.6 estimated plasma volumes) of FFP replaced per session, followed by a tapering course of prednisoneAFLP vs HSV hepatitis associated ALFAfter 3 d of TPE the patient’s mental status had returned to normal
Treatment with TPE was followed by biochemical and clinical improvement but during her recovery herpes simplex virus type 2 (HSV‐2) infection was diagnosed serologically and confirmed histologically
Shen et al[18]Case reportPE, No comparative armPlasma exchange: performed on days 1, 3, and 5, with 3000 mL of plasma exchanged during each sessionOccupational Exposure to TetrachloroethyleneBilirubin, ammonia, and prothrombin time improved before hospital discharge and patients mental status gradually became normal discharged on day 26 of hospital admission
Pashaei et al[30]Case reportPE, No comparative armPlasma exchange volume 2.5LWilson’s disease36 h after initiation of PE, encephalopathy recovered and there was no renal impairment. Copper, LDH total bilirubin decreased after the treatment
Plasma exchange fluid: FFP
Table 3 Studies included in for study of plasmapheresis in acute liver failure in pediatric cohort
Ref.Type of study / No. of patients recruitedStudy group(s)Plasma exchange regimeEtiology; AgeResults
Pham et al[19]Case seriesPE, No comparative armPlasma volume: Targeted 1-1.25 plasma volumesWilsons DiseasePost TPE 9 patients underwent liver transplantation and all 10 patients had at least 6 mo survival
n = 10
Age 6-61 yr
Median days from first to OLT was 1-53 d
Plasma exchange fluid: 77% of procedures were performed with plasma as sole replacement fluid while 23% used the combination of plasma and 5% albumin
Median number of TPE: 3.5
Chien et al[22]Retrospective case seriesPE, No comparative armPlasma exchange volume: Plasma exchange was usually performed daily for the first 3 d, and then shifted to every other day or every 3 d according to the patient's condition60% idiopathic, 17% infection, 8% metabolic and immunologic, 4% toxin11 (48%) had native liver recovery (NLR), 9 (39.1%) died without liver transplant, and 3 (12.9%) received liver transplantation
The no liver recovery group showed a lower proportion of idiopathic cases, lower peak ammonia level, higher peak alpha fetoprotein (AFP) level, and they had plasma exchange fewer times than the other groups
n = 23
Age 0.29-9.25 yr
Plasma exchange volume: 2–4 times the patient’s estimated plasma volume
Plasma exchange fluid: FFP
Ide et al[47]Retrospective case seriesPE/CVVHDF, No comparative armCVVHDF and PE were applied in all ALF patients2/17 viralAll laboratory results relating to liver dysfunction decreased significantly after CVVHDF + PE
1/17 mitochondrial
PE using 100 mL/kg of FFP per treatment course was implemented once daily for 6 to 8 h until the recovery of coagulopathyOverall survival rate 88% with median follow up period of 28 mo
14/17 indeterminate
n = 17
Age 1-11 mo [Median Weight 8.0 (2.7-10 kg)]
Verma et al[23]Case reportPE, No comparative armPlasma exchange volume: 1.5-2 h, 1.2 L plasma exchange in each session, in addition to oral D penicillamine and ZincWilson’s diseasePatient improved initially but subsequently deteriorated fter developing renal failure and shock, and died from acute pulmonary hemorrhage.
Age 5 yr (Weight: 15 kg)
Morgan et al[25]Case reportPE, No comparative armPlasma exchange volume: 1500 mL TPE, 5 single plasma volume over 11 d in addition to trientineWilson’s diseasePatient had worsening bilirubin, coagulopathy despite treatment and underwent OLT 12 d after beginning TPE
Plasma exchange fluid: PlasmaAge 6 years
Zhang et al[24]Case reportPE, No comparative armPlasma exchange volume: 1200 mL each time, with blood flow velocity of 45–50 mL/min, plasma separation speed of 650–750 mL/h, and a replacement time of approximately 2 h. Total 9 exchangesWilson’s diseaseCPFA started after PE. The patient had rapid recovery of consciousness, removal of copper and stabilization of serum bilirubin and hemoglobin. 9 d after last PE patient underwent liver transplant.
Plasma exchange fluid: FFPAge 7 yr (Weight 21 kg)
Yukselmis et al[26]Case reportPE, No comparative armPlasma exchange volume: 1.5 times total blood volume then 1 time for the subsequent coursesViral (Influenza)Patient did not require transplantation in light of clinical improvement and PE resulted in complete recovery
Total 3 sessions PE on top of ostelmavirAge 4 yr (Weight 16 kg)
Plasma exchange fluid: FFP
Ponikvar et al[27]Case reportPE+HD, No comparative armPlasma exchange volume: 3 volumes of plasma (12% of body weight of 16 kg) per procedure were exchanged (1972 ± 85 mL; range, 1800–2150 mL). FFP was used as the replacement solution. An equal volume of plasma was removed and replacedUnknown Excluded viruses and metabolic causePatient did not improve after 1 mo and was referred to a liver transplant center and successfully transplanted. Patient also had hyperbaric oxygen (HBO) during treatment
Age 3 yr (Weight 16 kg)
A total of 13 PEs, 13 HD sessions, and 9 HBO treatments over a period of 1mo. The initial 4 PEs were followed by HD sessions while the other 8 PE treatments were given simultaneously with HD. There was no renal failure; HD was instituted to improve ammonia elimination
Harmanci et al[48]Case report / Letter to editorPE, No comparative armPlasma exchange volume: 2.5 L per sessionWilson’s DiseasePatients mental status improved and was extubated and weaned from mechanical ventilation on the fifth day of hospitalization. The patient did not require liver transplantation. The patient was treated continously with zinc and D-penicillamine
Age 17 yr
Started daily and continued for 7 consecutive days
Table 4 Studies included for study of plasmapheresis in acute-on-chronic liver failure
Ref.Type of study/Number of patients recruitedStudy group(s)Characteristics of study populationPlasma exchange regimeEtiologyResults
Meng et al[39]Retrospective cohort study, single center; n = 158; PE group: 38; SMT group: 120PE vs SMTPE group: Higher MELD scorePerformed twice a week until patients’ condition was stable, additional weekly or biweekly visits were instituted if patients felt deterioration of their condition. Total duration of therapy 2-8 wkHepatitis B24/38 (63%) death in the PE group and 82% in SMT group died within 4 wks. By week 12, 71% in PE group and 86% in SMT group died
Baseline characteristics both groups had 26%-28% of patients with hepatic encephalopathy
ACLF definition: ACLF was defined as serum bilirubin ≥ 5 mg/dL and an INR 1.5 or prothrombin activity (PTA) 40 %, complicated within 4 wk by ascites and/or encephalopathy in patients with previously diagnosed or undiagnosed chronic liver diseases18% vs 14% transplant free survival in 3 mo comparing PE vs SMT (P < 0.01)
Plasma exchange volume not mentioned
Biochemically, there is decreased bilirubin in PE arm cf SMT (P < 0.01)
Mao et al[38]Retrospective cohort study, single centerPE vs SMTACLF definition: Acute decompensation of liver function in patients with chronic preexisting liver diseases. ACLF is defined as a syndrome with severe jaundice (total bilirubin: 171 mmol/L), coagulopathy (prolonged prothrombin time, prothrombin activity 40%), or hepatic encephalopathy (above grade II)Plasma exchange volume: 3500 mL at 25-30 mL/min. A total of 3000–4500 mL of fresh frozen plasma (40-60 mL/kg) and 20-40 g of human albumin were suppliedHepatitis B. Drug hepatitis, Wilson disease, alcoholic liver disease, autoimmune hepatitis excluded26 survivors and 36 non-survivors were in the PE group, whereas 33 survivors and 98 non-survivors were in the control group after 30 d treatment. Their survival rates were 41.9% and 25.2% for PE and medical therapy, respectively (P < 0.05)
Baseline characteristics 74%-77% had HE at baseline
PE group: 62
SMT group 131
No mention re: Biochemical improvement
Flow rate of blood was adjusted to 60–130 mL/min
Not randomised
PE was carried out 2-3 times per week for the first two weeks, then weekly, then stopped based on clinical results
Chen et al[42]Retrospective cohort study multicentre (10)PE, no comparative armACLF definition: Guidelines for Diagnosis and Treatment of Liver Failure in China (2006): Early stage is defined as a progressively deepening jaundice (Bilirubin level ≥ 171 μmol/L or a daily increase of ≥ 17.1 μmol/L), PTA > 30% but ≤ 40%, and no HE or other complications. Middle-stage disease represents progression of the symptoms of the early stage, including one of the following symptoms: Grades I/II HE, ascites, or a PTA of > 20% but ≤ 30%. In the end-stage disease, the condition deteriorates further with a PTA of ≤ 20% and includes one of the following symptoms: Hepatic-renal syndrome, severe upper gastrointestinal bleeding, serious infection, and grades III/IV HEPlasma exchange volume: 2500-3500 mL, and the PE rate was 20-25 mL/minHepatitis BForty-two of the 52 (80.8%) patients in the early stage, seventy-five of the 99 (75.8%) patients in the middle stage and thirty-seven of the 99 (37.4%) patients in the end stage survived for one month after diagnosis
n = 250 patients
Dexamethasone (5 mg) and heparin (2500 U) were injected routinely before PE
Authors concluded that late stage ACLF might benefit from PE as a bridge to definitive treatment-liver transplant
PE was repeated every 2-4 d
Zhou et al[45]Retrospective, cohortPBA + PE vs PEACLF was defined as acute liver injury emerging as jaundice and coagulopathy, complicated by ascites and/or encephalopathy within 4 wk in a patient with known or unknown chronic liver disease. The definition of liver failure in ACLF was as follows: Severe jaundice (total serum bilirubin ≥ 5 mg/dL) and coagulopathy (INR ≥ 1.5 or prothrombin activity < 40%) and ascites and/or encephalopathy.Plasma exchange volume: Approximately 3000 mL of plasma was exchanged per time at a blood flow rate of 20 to 25 mL/minHBV 56.6%, HBV+HEV 31.9% Others 11.5%The mean overall survival for the derivation cohort was 441 d (95%CI: 379-504), and the 90- and 270-d survival probabilities were 70.3% and 58.3%, respectively
Derivation cohort 113
Validation cohort 68The mean survival times of patients treated with PBA plus PE and patients treated with PE were 531 days (95%CI: 455-605) and 343 d (95%CI: 254-432), respectively (P = 0.012)
From the derivation cohort: PE, n = 54; PE+PA, n = 59
Each patient in the derivation cohort received PE 1 to 4 times
Predictors of survival: Age, MELD, Complication, type of ALSS No mention re: Baseline characteristics of PE vs PBA
Baseline population in this study is 51% cirrhotic
Wan et al[44]Prospective cohort studyPE vs DPMASACLF was defined as serum bilirubin 5 mg/dL and INR > 1.5 or PTA < 40%, complicated within 4 wk by ascites and/or HE in patients with previously diagnosed or undiagnosed chronic liver diseasesPlasma exchange volume: About 3000 mL of plasma exchanged at an exchange rate of 20-30 mL/min at each session.HBVDuring the study, a total of 42 patients died, with 24 in TPE group and 18 in DPMAS group. The median survival times were 12 wk in TPE group and 11 wk in DPMAS group
n = 60
TPE 33PE was performed 2-3 times/week, lasting 2-3 h every sessionThe 4-wk and 12-wk survival rates in TPE group and DPMAS group were 87.9% and 88.9%, 34.6% and 33.3%, respectively. There was no marked difference in survival between the two groups
DPMAS 27Baseline eAg positive greater in TPE group (18% vs 7.4%)
Bilirubin removal in TPE more efficient compared to DPMAS
Qin et al[37]Open label randomized controlled parallel group single-center studyPE centered ALSS vs SMTDefinition of ACLF was according to the Chinese guidelines, Bilirubin ≥ 10 mg/dL, PTA ≤ 40% and cirrhosis and multiorgan failure were not taken as mandatory criteria, according to the Chinese guidelinesPE volume: 3500 mL (40–60 mL/kg) FFP, at 25-30 mL/minHBVSurvival rates after 90 d were 60% (62/104) in ALSS-treated patients and 47% (61/130) in the control group. (P < 0.05). The 5-year cumulative survival rates of the ALSS and control groups were 43% (45/104) and 31% (40/130), respectively (P < 0.05)
ALSS schedule: 3 routine treatments were performed in the first 10 d after inclusion in the study (once per 3–4 d); extra treatments were offered according to the improvement of the patients. The methods of PE-centered ALSS were chosen based on clinical conditions. For patients with coagulopathy, PE was applied; for patients with encephalopathy, PE plus hemoperfusion or continuous hemodiafiltration was used; for patients complicated with HRS or imbalance of water or electrolytes, PE plus continuous hemodiafiltration was used
n = 234
No mention of biochemical improvement
Xia et al[40]Retrospective cohort studyNBAL (all had PE) vs SMTACLF definition:All of the patients were treated with PE, and most were treated with one or more additional methods, including 13/26 (50.00%) ALF patients, 16/27 (59.26%) Subacute ALF patients, and 228/407 (56.02%) ACLF patientsFor ACLF: 91.24% chronic hepatitis B, 3.69% alcohol abuse, 1.01% autoimmune, 1.01% cholestasis, 3.05% other causesClinical outcomes were improved after NBAL treatment. The 30-d survival rates of subacute liver failure (SALF) patients were 63% among those who received NBALs and 21% among those who did not receive NBALs (P < 0.01)
1 Acute deterioration of pre-existing chronic liver disease
n = 882
460 NBAL 422 control2 Extreme fatigue with severe digestive symptoms, such as obvious anorexia, abdominal distension or nausea and vomiting
The 30-day survival rate of acute-on-chronic liver failure (ACLF) patients who received NBALs was 47%, significantly higher than that of the non-NBAL patients (P < 0.05)
Of which 49 ALF, 46 SALF and 787 ACLF
3 Progressively worsening jaundice within a short period (serum total bilirubin ≥ 10 mg/dL or a daily elevation ≥ 1 mg/dL)
The choice of therapy was based on each patient’s condition: PE in combination with PP for HE was administered in 12.24% (6/49) of ALF patients, 10.77% (7/65) of SALF patients, and 7.41% (80/1079) of ACLF patients. In patients with HRS, we administered PE with CHDF in 32.65% (16/49) of ALF patients, 23.08% (15/65) sessions of SALF patients and 28.17% (304/1079) sessions of ACLF patients
Reported to be effective in biochemical improvement
4 Obvious hemorrhagic tendency with PTA ≤ 40% (PT ≥ 18.3 s or INR > 1.50)
The absence of any of the above four criteria precluded a diagnosis of ACLF
Pts underwent 1-4 times of NBAL each
Li et al[49]Prospective cohort StudyPE vs PE + UCMSCsACLF was defined as serum bilirubin ≥ 5 mg/dL and INR ≥ 1.5 or PTA < 40%, complicated within 4 wk by ascites and/or encephalopathy in patients with previously diagnosed or undiagnosed chronic liver diseasePE volume: About 3000 mL, and the exchange rate of plasma was 20–30 mL/min. Heparin was used as anticoagulant during PEHBVThe cumulative survival rates at 3 mo in group A and group B were 54.5 % and 29.4 %, respectively (P = 0.015 by log rank test)
PE: 34
INR was prominently lower in PE + UCMSCs group than in PE group (P < 0.05). At 12 mo, patients in PE+UCMSCs group showed lower levels of AST than patients in PE group (P < 0.05).
PE+UCMSCs: 11
n = 45
In PE group: MELD score: 22.5 +/- 1.4, 61.8% cirrhoticAt 24 mo, patients in PE+UCMSCs group had significantly improved levels of albumin, PT and INR than patients in PE group (P < 0.05). However, ALT, Total bilirubin, Direct bilirubin, creatinine, white blood cell, Hemoglobin, Platelet and ascites were comparable at each follow-up
Xu et al[43]Retrospective cohort studyPEDefinition of ACLF: Acute hepatic insult manifesting as jaundice and coagulopathy, complicated within 4 wk by ascites and/or encephalopathy in a patient with previously diagnosed or undiagnosed chronic liver diseaseTotal volume exchanged 3300 mLHBV1-yr and 5-yr survival rates in the ALSS-LT group and LT group were 79.2% and 83%, 69.7% and 78.6%
n = 171
Patients with coagulopathy were indicated for PE, when the patient had HE, PE + hemodiafiltration was used. For patient with hepatorenal syndrome or imbalance of water or electrolytes, PE + continuous hemodiafiltration or MARS was used
PE before LTx: 115
Emergent LTx: 56
PE group: MELD score 31+/-6
Yao et al[41]Retrospective cohort studyPE vs DPMAS + PEDefinition of ACLF: Acute hepatic insult manifesting as jaundice and coagulopathy, complicated within 4 wk by ascites and/or encephalopathy in a patient with previously diagnosed or undiagnosed chronic liver diseasePE volume: Fresh frozen plasma was 2200 to 2400 mL per treatment. Duration of single treatment was about 2 hHBVThe total bilirubin levels immediately after treatment at 24 and 72 h after treatment were markedly decreased in DPMAS + PE group compared to that in PE group (52.3 ± 9.4% vs 42.3 ± 7.2%, P < 0.05; 24.2 ± 10.0% vs 13.5 ± 13.0%, P < 0.05; 24.8 ± 13.1% vs 14.9 ± 14.9%, P < 0.05; respectively).
n = 131
PE group (n = 77)Patients underwent 1-4 times of PE / PE + DPMAS
DPMAS + PE group (n = 54)Baseline characteristics were similar in both groups
The 28- d survival rates was 62.3% and 72.2% in PE and DPMAS + PE groups (P = 0.146).
28- d survival rates were significantly higher in DPMAS + PE group than that in PE group (57.4% vs 41.7%, P = 0.043) in the intermediate-advanced stage patients
Cheng et al[12]Retrospective, cohort study single tertiary centrePE, no comparative armACLF definition: acute hepatic insult that manifests as jaundice (serum bilirubin ≥ 5 mg/dL and coagulopathy (INR ≥ 1.5), which is complicated within 4 wk by clinical ascites and/or encephalopathy in a patient with previously diagnosed or undiagnosed chronic liver disease or cirrhosisThe processed plasma volume was approximately 3000 mL for each session (1-1.5 total plasma volume); the blood flow rate was 100 mL/min; and the PE rate was 25–30 mL/min, with an equivalent volume of replacement fluid using fresh frozen plasmaHepatitis B (75%) in ACLF group, 6% alcohol, others: HCV, AIHBiochemical improvements seen after PE: AST/ALT/Bil/INR
Average 4-5 sessions of PE in ACLF group, 3-5 sessions in ALF
n = 55; 10 ALF, 45 ACLF
Initial diagnosis to PE is longer in non-survivors in ACLF and ALF though not significant
Survival based on etiology of ACLF: 24% HBV, 67% ETOH, 0% HBV + alcohol, 0% HCV 0% HCV + alcohol, 0% AIH
79% of patients with ACLF have cirrhosis, 55% have grades III-IV HEPE occurred daily or every other day till sustained clinical improvement, liver transplantation or no clinical response/death
Table 5 Studies included for use of plasmapheresis in acute liver failure and acute-on-chronic liver failure in adults
RefType of study/No. of patients recruitedComparative armPlasma exchange regimeEtiologyResults
Xia et al[40]n = 882; 460 NBAL 422 control; Of which 49 ALF, 46 SALF and 787 ACLFNBAL (all had PE) vs SMTAll of the patients were treated with PE, and most were treated with one or more additional methods, including 13/26 (50.00%) ALF patients, 16/27 (59.26%) SALF patients, and 228/407 (56.02%) ACLF patients.For ACLF: 91.24% chronic hepatitis B, 3.69% alcohol abuse, 1.01% autoimmune, 1.01% cholestasis, 3.05% other causesClinical outcomes were improved after NBAL treatment. The 30-d survival rates of subacute liver failure (SALF) patients were 63% among those who received NBALs and 21% among those who did not receive NBALs (P < 0.01)
The choice of therapy was based on each patient’s condition: PE in combination with PP for HE was administered in 12.24% (6/49) of ALF patients, 10.77% (7/65) of SALF patients, and 7.41% (80/1079) of ACLF patients. In patients with HRS, we administered PE with CHDF in 32.65% (16/49) of ALF patients, 23.08% (15/65) sessions of SALF patients and 28.17% (304/1079) sessions of ACLF patientsFor ALF: 42% drug toxicity, 16% HBV, 10% surgical trauma, 30% unexplainedThe 30-day survival rate of acute-on-chronic liver failure (ACLF) patients who received NBALs was 47%, significantly higher than that of the non-NBAL patients (P < 0.05)
Pts underwent 1-4 times of NBALFor SALF: 54% drug toxicity, 30% unexplained, 4% Hepatitis E, 11% HBVReported to be effective in biochemical improvement
Cheng et al[12]Retrospective, cohort study single tertiary centre; n = 55; 10 ALF, 45 ACLFPE, no comparative armPE volume: About 3000 mL, and the exchange rate of plasma was 20-30 mL/min. Heparin was used as anticoagulant during PEIn ALF group: 50% HBV, 20% drug, others include ischemic hepatopathy, traumatic liver injury, HLH20% (1/5) of the HBV related ALF survived, 1/2 of drug related ALF survived, and 1/1 of the traumatic liver injury related ALF survived.
Significant improvements see in levels of serum total bilirubin, AST ALT INR PT. No significant changes in ammonia
Nakae et al[21]Retrospective case series; n = 21; 10 FH; 11 ALFPDF, no comparative armPE volume: 1200mL of normal FFP and 50mL of 25% albumin solution was infused intravenously over 8 hFH90 d survival:
70% Hep B20% in FH patients
10% AIH54.5% in ALF patients
20% DrugOverall survival 38.1%
The PDF session lasted 8h, and the blood flow rate was 100 mL/min. Filtered replacement fluid for was infused at a dialysate flow rate of 600 mL/h and a replacement flow rate of 450 mL/h
Lower MELD correlated to increased survival
ALF
No patients survived beyond 90 d with MELD > 40
Biochemically: Bilirubin, IL-18 statistially different when compared before and after PDF
3/11 Unknown
1/11 GVHD
4/11 ETOH
1/11 HBV
Fluid removal was performed by reducing the replacement flow rate to 450 mL/h at most1/11 EBV
1/11 Drug
5/11 was labelled as AOCLF
Pu et al[34]Case series (excluding patients who abandoned treatment; n = 33); 8 ALF; 3 SALF; 14 ACLFCHDF followed by sequential PE, No comparative armPatients underwent continuous hemofiltration on a daily basis during the daytime followed by sequential treatment with plasma exchange 1800-2400 mL or hemodialysis every 2-3 d29 patients with hepatitis B virus infection, 1 with Hepatitis E virus infection, and 3 patients with unknown etiology; 18 were male and 15 female; age ranged from 23 to 65Restoration of consciousness in 6 of 8 cases (75%) in acute liver failure (ALF) group, 3 of 3 cases (100%) in subacute liver failure (SALF) group, and 9 of 14 cases (64.29%) in acute/subacute on chronic liver failure (A/SCLF) group
Of all cases, 11 patients restored consciousness after 7 d in a coma. The rate of long-term survival (those who abandoned the treatment were excluded) was 3/7 (42.86%) for ALF group, 2/2 (100%) for SALF group, and 1/11 (9.09%) for A/SCLF group
No mention of biochemical changes
Schaefer et al[50]Retrospective cohort study; n = 10; 8 had combined PE, HD + MARSPE + HD + MARS vs MARSPE volume: 1.5 plasma volume was exchanged per session within 2–3 hWilson’s disease in 2 patients, congenital liver fibrosis, progressive intrahepatic cholestasis, severe combined immunodeficiency, disseminated herpes simplex virus 2 infection, multi-organ failure due to mycoplasma-induced myocarditis, autoimmune hepatitis, fungal sepsis and cetirizine intoxicationMARS and PE/HD treatments were well tolerated by all patients. No bleeding episode occurred. 1 patient with multi-organ failure due to mycoplasma-induced myocarditis, 1 with cetirizine intoxication completely recovered. 3 patients were successfully transplanted, five children died with multi-organ failure and sepsis, including the three children treated with Mini-MARS
PE was immediately followed by a HD session in six children, using the same extracorporeal circuit with a polysulfone high-flux filter (Fresenius)
Standard MARS treatment only slightly decreased serum bilirubin (16.3 ± 6.5-13.8 ± 5.9 mg/dL) and ammonia (113 ± 62-99 ± 68 μmol/L) and international normalized ratio (INR) tended to increase (1.5 ± 0.3 and 2 ± 1.1)
2 had MARS only
Mini-MARS did not reduce serum bilirubin, ammonia slightly decreased and INR increased
Age 0.1-18 yr
PE/HD reduced serum bilirubin (23 ± 8.4-14.7 ± 7 mg/dL), ammonia (120 ± 60–70 ± 40 μmol/L) and INR (2.4 ± 0.8-1.4 ± 0.1, all P < 0.05). Intraindividual comparison showed a slight increase in bilirubin by 2 ± 22% with MARS and a reduction by 37 ± 11% with PE/HD (P < 0.001 vs MARS) and a decrease in ammonia of 18% ± 27% and 39% ± 23% (P < 0.05). INR increased during MARS by 26 ± 41% and decreased with PE/HD by 37 ± 20% (P < 0.01)
Singer et al[51]Retroespective case seriesNo comparative arm, TPE in all patientsPlasma volume removed per exchange was 121 ± 47 mL/kg (2.2 ± 0.6 plasma volume) of FFP57% FHF, 18% BA, 20% IEM, 5% other of note 43% had CLDCoagulation profiles after TPE significantly improved compared with mean pre-exchange values
Spontaneous recovery was observed in three patients; the remaining either underwent transplantation (32/49) or were not considered transplant candidates because of irreversible neurologic insults (11/49) or sepsis (3/49)
Age 10 d to 18.4 yr