Emergency adult-to-adult living-donor liver transplantation for acute liver failure in a hepatitis B virus endemic area


  • Potential conflict of interest: Nothing to report.


The outcomes of patients with acute liver failure (ALF) vary greatly according to etiology. Emergency adult-to-adult living-donor liver transplantation (adult LDLT) would help address the shortage of available organs for patients with ALF, especially in hepatitis B virus (HBV)-endemic areas. We analyzed a prospective database of 110 consecutive adult patients with ALF. ALF was defined as sudden development of severe coagulopathy and encephalopathy within 26 weeks of onset of symptoms. In about 90% of patients, ALF was caused by etiologies that usually result in poor outcomes, including HBV infection (37%). Three cases (3%) were associated with acetaminophen overdose. Of the 99 patients listed for emergency liver transplantation, four (4%) underwent deceased-donor liver transplantation (DDLT), and 40 (40%) underwent adult LDLT. The 1-year survival rate of adult LDLT patients was 85%. Of the 55 patients listed but not transplanted, 45 (82%) died within a median of 7 days (range, 1-90 days). Multivariate analysis showed that adult LDLT (hazard ratio [HR] 0.10, P < 0.01) and DDLT (HR 0.12, P = 0.04) were associated with decreased mortality, whereas older age (HR 1.03, P = 0.01) and higher Model for End-stage Liver Disease (MELD) (HR 1.03, P = 0.04) was associated with increased mortality of patients. There was no living donor mortality. Eight (17.8%) and three (6.7%) living donors experienced grade 1 and 2 complications, respectively. Conclusion: Emergency adult LDLT can be performed expeditiously and safely for patients with ALF, and greatly improves the survival rate. As the window during which transplantation is possible is limited, emergency adult LDLT should be considered one of the first-line treatment options in patients with ALF, especially in regions in which ALFs are caused by etiologies associated with poor outcome and the supply of organs is severely limited. (HEPATOLOGY 2010.)

Acute liver failure (ALF) is a condition in which rapid deterioration of liver function results in altered mentality and coagulopathy in individuals without preexisting cirrhosis.1 The probability of spontaneous recovery is usually poor, with emergency liver transplantation (LT) often being the only effective treatment.1 The cause of ALF is the most important determinant of patient outcomes.2 For example, spontaneous recovery rates from ALF caused by hepatitis B virus (HBV) infection are significantly lower than from ALF attributable to acetaminophen (APAP) toxicity.3 The etiology of ALF varies markedly by geographical region.2

Because ALF progresses rapidly, the need for LT is urgent. In Western countries, not all patients listed for LT receive a liver graft from a deceased donor, and the death rate of patients awaiting LT ranges from 10% to 40%.3, 4 The donation rate from deceased donors is even much lower in Asian countries.5, 6 The high mortality rates from ALF and the limited number of organs available from deceased donors has led to the use of adult-to-adult living donor liver transplantation (adult LDLT) in many countries.7–9 Emergency adult LDLT is likely to be particularly effective for patients in regions where ALF is mainly caused by etiologies associated with a high mortality rate and a severely limited supply of organs. However, its use has been limited by ethical concerns and the time constraints needed to evaluate donors.

In this prospective cohort study we evaluated the long-term effect and donor safety of emergency adult LDLT for ALF in Korea, an HBV-endemic area with a severe shortage of organs from deceased donors.


AIH, autoimmune hepatitis; ALF, acute liver failure; APAP, acetaminophen; BMI, body mass index; CHB, chronic hepatitis B; CMV, cytomegalovirus; CNI, calcineurin inhibitor; CT, computed tomography; DDLT, deceased-donor liver transplantation; EBV, Epstein-Barr virus; GRWR, graft-recipient weight ratio; GV, graft volume; HBV, hepatitis B virus; HR, hazard ratio; INR, international normalized ratio; IQR, interquartile range; KONOS, Korean Network for Organ Sharing; LDLT, living-donor liver transplantation; LT, liver transplantation; LT group, patients who received liver transplants; MELD, Model for End-stage Liver Disease; no-LT group, patients listed but not transplanted; RUCAM, Roussel Uclaf Causality Assessment Method; SAE, severe acute exacerbation; SD, standard deviation; SLV, standard liver volume; USN, ultrasonography.

Patients and Methods

Study Subjects.

We analyzed a prospective database of all adult patients consecutively admitted to the liver unit of our institution between January 2004 and June 2007. ALF was diagnosed according to the criteria of the American Association for the Study of Liver Diseases (AASLD),1 including a sudden development of severe coagulopathy with an international normalized ratio (INR) ≥1.5 and mental alteration with an illness duration no longer than 26 weeks. Patients with underlying chronic diseases such as chronic hepatitis B (CHB) and autoimmune hepatitis (AIH) were included if they had normal liver function before the onset of symptoms and there was no evidence of cirrhosis. Informed consent was obtained from each patient's next of kin, and the study was approved by the Institutional Review Board of the Asan Medical Center.

Etiologic Diagnosis.

All patients underwent standardized evaluation to determine the cause of liver injury. This included a detailed review of medications, herbal remedies, and exposure to toxins; assays for hepatitis B surface antigen (HBsAg), anti-hepatitis B core (IgM), anti-hepatitis A virus (IgM), anti-hepatitis C virus, and anti-hepatitis E virus (IgM) antibodies; serologic tests with or without polymerase chain reaction (PCR) quantification of cytomegalovirus (CMV), Epstein-Barr virus (EBV), herpes simplex virus, and human immunodeficiency virus; assays for antinuclear and anti–smooth muscle antibodies; serum ceruloplasmin and 24-hour urine copper quantification; and liver imaging, including dynamic contrast-enhanced computed tomography (CT).

Drug-induced or herb-induced ALF was diagnosed when a temporal relationship between exposure to a suspected agent and the onset of ALF was identified and other causes of ALF were excluded. APAP was presumed as the cause of ALF when there was a history of potentially toxic APAP ingestion (>4 g/day) within 7 days of presentation. Patients were diagnosed as having ALF attributable to other drugs or herbs by use of the Roussel Uclaf Causality Assessment Method (RUCAM)10 or a locally developed scale for assessment of phyto-hepatotoxicity (a modified RUCAM).11 AIH was diagnosed using the criteria of the International Autoimmune Hepatitis Group.12 ALF was classified as “indeterminate” when supporting evidence of a specific etiology could not be established despite extensive investigation.

Evaluation for Adult LDLT.

Evidence of irreversible brain damage, uncontrolled sepsis, severe irreversible cardiopulmonary disease, extrahepatic malignancy, or advanced age (>70 years) was considered a contraindication for transplantation.

Immediately following the diagnosis of ALF, patients without contraindications were listed on the Korean Network for Organ Sharing (KONOS) and were given national priority (status 1) for available deceased-donor livers. At the same time, the need for an emergency LT was explained to each patient's next of kin, who were also informed in detail of the risks and benefits of deceased-donor liver transplantation (DDLT) and adult LDLT. Maximum efforts were made to avoid any coercion and written informed consent was obtained from each living donor candidate according to guidelines of the Institutional Ethics Committee. The spontaneous willingness of each potential donor was confirmed by social workers, transplantation coordinators, and psychologists if necessary. All donations were approved by the Institutional Ethics Committee and KONOS. Evaluation of a living-donor candidate, however, did not preclude or delay DDLT if a suitable deceased-donor liver became available during living-donor evaluation.

Living-donor candidates were admitted to the emergency room for donor evaluation and all procedures were performed in an emergency manner. Living donors were selected on the basis of complete medical history, physical examination, laboratory findings, imaging data including abdominal ultrasonography (USN), CT for graft/recipient size matching (three-dimensional CT with volumetric analysis), and routine percutaneous USN-guided liver biopsy. The degree of steatosis was immediately evaluated by a pathologist using frozen sections of the liver biopsy. Donor candidates in whom liver histology showed >30% steatosis were not accepted. ABO-blood groups were identical or compatible in all cases. The minimally required graft volume to ensure metabolic demands of patients was an estimated graft-recipient weight ratio (GRWR) ≥0.8 or an estimated graft volume (GV) ≥40% of the standard liver volume (SLV). When a single-graft transplant did not appear feasible after consideration of donor safety (remnant volume <30% of total liver volume and/or severe steatosis) and the possibility of a small-for-size graft for the recipient, a dual-graft transplant was considered as a last resort.

Post-LT Management of Recipients and Donors.

The peritransplantation primary immunosuppression protocols used for recipients of both deceased and living donor organs consisted of interleukin-2 receptor inhibitor (basiliximab) on days 0 and 4; an intraoperative bolus of methylprednisolone (5-10 mg/kg); intravenous or oral calcineurin inhibitor (CNI), such as cyclosporine or tacrolimus, with corticosteroid recycling beginning on day 1; and adjunctive mycophenolate mofetil for patients showing CNI-associated side effects or suspected mild or acute cellular rejection. Corticosteroid was rapidly tapered within the first 3 months. Immunosuppression was not reduced for patients with HBV-associated ALF.

All patients who underwent transplants because of HBV infection received long-term passive anti-HBs immunoprophylaxis to maintain the serum anti-HBs titer above 500 IU/L. Oral nucleos(t)ide analogs, including lamivudine, adefovir, or entecavir, were used to treat HBV recurrence. Liver recipients were followed in the outpatient clinic every 2 weeks for the first 3 months and as clinically indicated thereafter.

All living donors were closely monitored in the intensive care unit for the first 1 or 2 days after donation, especially for early, timely detection of bleeding. Follow-up CT scans were performed 1 week, 1 month, 3 months, and 1 year after surgery, according to customized protocols.

Statistical Analysis.

Patient survival was analyzed from the time of diagnosis. Patients and living donors were followed until death or the end of August 2009. Information on these outcomes was taken from patient registration data and medical records. The primary endpoint of this study was 1-year survival after diagnosis. Kaplan-Meier analysis was used to estimate overall survival rates and compared between groups using the log-rank test.

Potential prognostic factors for survival were evaluated at the time of diagnosis and were included in univariate and multivariate analyses using a proportional hazards model. These factors included recipient age; gender; etiology; days from jaundice to encephalopathy; grade of encephalopathy; presence of chronic liver disease; INR; alanine aminotransferase, aspartate aminotransferase, bilirubin, albumin, blood urea nitrogen, creatinine, serum sodium, and bicarbonate; arterial pH; white blood cell count; hemoglobin; platelet count; alpha-feto protein; the model for end-stage liver disease score (MELD); and GV/SLV. MELD was calculated according to the original formula: MELD = 11.2 LN(INR) + 3.78 LN(bilirubin) + 9.57 LN(creatinine) + 6.43. Days from diagnosis to LT; application of continuous hemodiafiltration; serum lactate and arterial ammonia concentrations immediately before LT; donor age, gender, and GRWR; and graft steatosis were also included in the analysis of factors predictive of post-LT survival. Missing data were not replaced.

All statistical analyses were performed using SPSS for Windows (SPSS, Chicago, IL). P < 0.05 was defined as statistically significant in all analyses.


Study Subjects.

A total of 110 patients were enrolled in this study (Fig. 1). Of these, 11 had contraindications for LT at the time of diagnosis (contraindications to LT group), including irreversible brain edema (n = 2), uncontrolled septic shock (n = 1), underlying malignancy (n = 7), and advanced age (n = 1). Thus, 99 patients were listed for transplantation, of whom 44 (44% of those listed, 40% of total study patients) underwent LT (LT group), with four undergoing DDLT and 40 undergoing adult LDLT. Five of these patients received dual grafts from two living donors each, including four who received two left-lobe grafts and one who received left- and right-lobe grafts. The remaining 55 patients did not undergo LT (no-LT group) because of the lack of a timely suitable donor.

Figure 1.

Flow chart showing the treatment of 110 patients with acute liver failure. ALF, acute liver failure; LDLT, living-donor liver transplantation; DDLT, deceased-donor liver transplantation; LT, liver transplantation. Contraindications to LT group: patients not listed for LT because of the presence of a contraindication; LT group: patients who received LT; no-LT group: patients listed but not transplanted.

The median age of the overall study cohort was 42 years (range, 17–74 years), with 56 (51%) being males (Table 1). The median duration of symptoms before the onset of encephalopathy was 10 days (interquartile range [IQR] 4–21 days), and the median interval between the onset of jaundice and encephalopathy was 5 days (IQR, 1–13 days). The encephalopathy grade at diagnosis was grade 1 or 2 in 96 (87%) patients and grade 3 or 4 in 14 (13%) patients. The median baseline MELD was 31.8 (IQR 25.6–39.3).

Table 1. Baseline Characteristics of Patients
CharacteristicsTotalContraindication to LTNo-LTLTP*
  • Contraindication to LT, patients with contraindications for liver transplantation; no-LT, patients listed but not transplanted; LT, patients who underwent liver transplantation.

  • In the LT group, 40 patients underwent living-donor, and four underwent deceased-donor liver transplantation.

  • HBV, hepatitis B virus; INR, international normalized ratio; MELD, model for end-stage liver disease.

  • *

    P value difference between the no-LT and LT groups

  • Duration before encephalopathy onset.

  • Median (Interquartile range, IQR).

  • §

    Median (range).

Age, years (median, range)42 (17–74)56 (25–74)44 (17–68)36 (18–59)<0.01
Gender (male)56, 51%7, 64%26, 47%23, 52%0.69
Etiology    0.42
 HBV41, 37%3, 27%22, 40%16, 36% 
 Herbs/Mushrooms26, 24%5, 46%9, 16%12, 27% 
 Others43, 39%3, 27%24, 44%16, 36% 
Symptoms (days)10 (4–21)7 (4–11)12 (6–22)7 (3–16)0.60
Jaundice (days)5 (1–13)2 (0–7)6 (1–17)5.5 (1–11)0.86
Encephalopathy at Diagnosis    0.76
 Grade 1–296, 87%9, 82%49, 89%38, 86% 
 Grade 3–414, 13%2, 18%6, 11%6, 14% 
Bilirubin (mg/dL)17.6 (9.0–27.3)10.9 (5.9–20.0)17.5 (7.6–25.2)22.0 (11.0–30.8)0.06
INR3.1 (2.4–4.6)2.8 (2.1–7.3)3.2 (2.2–6.2)3.1 (2.7–4.2)0.29
Creatinine (mg/dL)1.0 (0.7–1.8)1.1 (0.7–1.3)1.0 (0.7–2.1)1.0 (0.6–1.6)0.65
MELD31.8 (25.6–39.3)27.8 (24.0–32.8)32.7 (25.1–40.5)30.7 (26.8–38.0)0.66
Arterial pH7.48 (7.43–7.52)7.45 (7.37–7.51)7.48 (7.42–7.51)7.49 (7.44–7.52)0.17
Hemodiafiltration36, 33%4, 36%21, 38%11, 25%0.20
Follow-up period (days)§36 (1–1989)6 (1–67)9 (1–1841)1100 (4–1989)

The decision to perform emergency LT was based on the progression of encephalopathy to grade 3 or 4 and on the availability of a suitable liver graft. Most baseline demographic and laboratory features did not differ significantly between the LT and no-LT groups (P > 0.05), except that median age was significantly younger in the LT group (P < 0.01).

Causes of ALF.

Overall, HBV was the most common cause of ALF (Fig. 2), accounting for 41 cases (37%). Of these, 14 (34%) were caused by acute HBV infection, whereas 27 (66%) were attributable to severe acute exacerbation (SAE) of preexisting CHB, either spontaneously (n = 21), by the development of a resistance mutation to lamivudine (n = 3), or after withdrawal of immunosuppressive therapy (n = 3). All patients with SAE of CHB conformed to the AASLD diagnostic criteria for ALF,1 and all had normal liver function before the onset of symptoms and no evidence of cirrhosis. Of the 41 patients associated with HBV, 38 were listed for LT. Of these, 30 patients were given lamivudine (n = 26) or entecavir (n = 1) as initial treatment, or lamivudine plus adefovir (n = 3) as salvage treatment for SAE of CHB associated with lamivudine resistance immediately after identifying the cause. The second most common etiology of ALF was herbal remedies used as complementary or alternative medicine, observed in 21 patients (19%). Most of these patients had ingested nonprescribed preparations containing multiple plants or herbs, making the identification of a single hepatotoxic herbal ingredient difficult. Other causes of ALF included acute hepatitis A (n = 8, 7%), AIH (n = 8, 7%), drugs other than APAP (antituberculosis agents, nonsteroidal anti-inflammatory drugs, valproic acid, and sevoflurane; n = 7, 6%), and mushroom poisoning (by Amanita virosa and A. subjunquillea; n = 6, 5%). APAP overdose was the cause of ALF in three patients (3%). Miscellaneous causes, observed in seven patients (6%), included dimethylnitrosamine ingestion, trichloroethylene exposure, EBV, CMV, veno-occlusive disease, and radiation overexposure. No cause was identified in 10% of patients, and these were classified as indeterminate ALF. There was no significant difference in the distribution of etiologies between the LT and no-LT groups (P > 0.05, Table 1).

Figure 2.

Causes of acute liver failure in 110 patients. HAV, hepatitis A virus; HBV, hepatitis B virus.

Seven patients had underlying malignancies, including three with advanced gastric cancer, two with non-Hodgkin's lymphoma, and one each with acute myeloid leukemia and chronic lymphocytic leukemia. The direct causes of ALF included herbal medication in two patients, SAE of CHB in two, veno-occlusive disease in one, extensive radiation-induced liver disease in one, and indeterminate in one.

Pathologic Findings on Explanted Livers.

The mean ± standard deviation (SD) weight of the 44 explanted livers in the LT group was 850 ± 378 g. There was no difference in mean weight between the 12 patients with SAE of CHB and the 32 other patients (870 ± 428 g versus 843 ± 366 g, P = 0.84).

Pathological examination of the explants showed massive or submassive necrosis in all patients and moderate to marked hepatitis in 33 patients (75%). Bridging fibrosis was observed in 23 patients (52.3%), and no or minimal fibrosis in the remainder. No patients had definite features of cirrhosis. The proportion of patients with bridging fibrosis did not differ significantly between patients with SAE of CHB and others (66.7% versus 46.9%, P = 0.24).

Patient Outcomes.

Overall patient survival was 42.7% (47 of 110 patients). All 11 patients with contraindications to LT died within 10 weeks of diagnosis (Fig. 3). Of the 55 patients in the no-LT group, 45 (82%) died while awaiting a graft, with a median time from diagnosis to death of 7 days (IQR 4-11 days). Of the 49 patients in the no-LT group who had grade 1 or 2 encephalopathy at enrollment, only six (12.2%) remained at encephalopathy grade 1 or 2, and all six recovered spontaneously. In contrast, 43 (87.8%) of these 49 patients progressed to grade 3 or 4, with only four (9.3%) recovering spontaneously. All of the survivors (n = 10, 18%) in the no-LT group recovered fully and maintained normal liver function after a median follow-up period of 1,277 days (range, 855–1,841 days). Among the 56 patients who died without transplantation, the most common cause of death was cerebral edema (46%) followed by infection (43%).

Figure 3.

Kaplan-Meier analysis of 1-year survival of patients with acute liver failure.

All patients in the LT group progressed to grade 3 or 4 encephalopathy before receiving LT. Four patients received liver grafts from deceased donors on days 2, 5, 6, and 10, respectively, after diagnosis. The median time from diagnosis to adult LDLT was 2.5 days (range, 0–26 days). The 1-year patient survival rate of the adult LDLT group was 85% (34 of 40 patients), significantly higher than that of the no-LT group (P < 0.01), but similar to that of the DDLT group (75%, P > 0.05; Fig. 3). The 1-year graft survival rate for the adult LDLT group was the same as the patient survival rate. Six adult LDLT patients, including one who received a dual-graft and one DDLT patient, died within 6 months as a result of brain edema (n = 2), systemic infection (n = 4), or bleeding (n = 1). One LDLT patient underwent a second transplantation for graft failure caused by acute cellular rejection, but later died of fungal pneumonia and sepsis. None of the 1-year survivors in the LT group (n = 37) died within a median follow-up period of 1,168 days (range, 465–1,989 days).

Prognostic Factors for 1-Year Survival.

Univariate analysis of the 99 patients listed for LT showed that adult LDLT was inversely associated with 1-year mortality (hazard ratio [HR] 0.09, P < 0.01; Table 2). Older age (HR 1.05, P < 0.01), higher INR (HR 1.08, P = 0.04), higher MELD (HR 1.03, P = 0.03), and lower arterial pH (HR 0.001, P = 0.01) were significantly associated with 1-year mortality. Multivariate analysis showed that adult LDLT (HR 0.10, P < 0.01) and DDLT (HR 0.12, P = 0.04) were independently associated with decreased mortality, whereas older age (HR 1.03, P = 0.01) and higher MELD (HR 1.03, P = 0.04) were independently associated with increased mortality.

Table 2. Significant Predictive Factors for 1-Year Mortality in Patients Listed for LT
 HR95% CIP
  • Subjects, n = 99; event, death within 1 year of diagnosis (n = 52). Variables measured at the time of diagnosis.

  • *

    Loge value.

  • Cox Proportional Hazards model with a backward elimination method.

  • Adult LDLT, adult-to-adult living-donor liver transplantation; CI, confidence interval; DDLT, deceased-donor liver transplantation; HR, hazard ratio; INR, international normalized ratio; MELD, model for end-stage liver disease.

Univariate analysis
Adult LDLT0.090.04–0.22<0.01
Arterial pH*0.0010.001–0.0040.01
Multivariate analysis
Adult LDLT0.100.04–0.26<0.01

In the LT group, significant factors predicting 1-year posttransplantation mortality were pretransplantation hemodiafiltration (HR 4.62, P = 0.05), higher creatinine level (HR 2.23, P = 0.02), lower arterial pH (HR 0.001, P = 0.03), and higher serum lactate concentration (HR 3.63, P = 0.04; Table 3).

Table 3. Significant Predictive Factors for 1-Year Posttransplantation Recipient Mortality, by Univariate Analysis
 HR95% CIP
  • Subjects, n = 44; event, recipient death within 1 year after LT (n = 7).

  • Variables were measured at the time of liver transplantation.

  • *

    Loge value.

Arterial pH*0.0010.001–0.0040.03

Characteristics and Complications of Living Donors.

In total, 72 living donor candidates for 48 patients underwent donor work-up. Of these, 35 were accepted as donors of single right-lobe grafts and 10 for dual-graft implantation. There were no ABO-incompatible donors.

Causes of 27 donor rejections included disproportionate future remnant left liver volume (n = 19), excessive steatosis (n = 3), failure to obtain permission from the Institutional Ethics Committee and KONOS (n = 3), HBsAg positivity (n = 1), and withdrawal of donation willingness (n = 1). No potential donors were rejected because of variations of donor vascular and biliary anatomy. The four patients who underwent DDLT had no potential living donors. Of the 55 patients in the no-LT group, 8 had 12 potential donors, who were rejected because of disproportionate interlobar liver volume proportions (n = 8), excessive steatosis (n = 2), HBsAg positivity (n = 1), or withdrawal of donation willingness (n = 1).

The 45 living donors were age 16 to 53 years (median, 27 years); 25 (56%) were female (Table 4). Of these, 42 (93%) were family members and three (7%) were emotionally motivated unrelated donors. Their median degree of hepatic steatosis was 5% (range, 0–30%); <5% in 33, 5%–25% in 11, and 25%–30% in one. The degree of donor hepatic steatosis was not associated with length of hospital stay, the occurrence of hepatic insufficiency, or any other donor complication (all P > 0.05). None of the donor or graft characteristics, including donor age, gender, GRWR, or graft steatosis, was associated with 1-year posttransplantation recipient mortality (all P > 0.05; data not shown).

Table 4. Donor Characteristics and Complications
  • *

    Data expressed as median (range).

  • Clavien classification.13

  • BMI, body mass index; ICU, intensive care unit.

Male, N (%)20, 44%
Age (years)*27 (16–53)
Relationship with recipient, N (%) 
 Family member42 (93%)
 Emotionally motivated unrelated donors3 (7%)
Type of graft, N (%) 
 Right lobe36 (80%)
 Left lobe9 (20%)
Steatosis*5% (0–30%)
BMI (kg/m2)*21.2 (17–28)
ICU stay (days)*2 (1–3)
Total hospitalization (days)*14 (9–24)
Mortality, N0
Complications,N (%)11 (24.4%)
 Grade 1, N (%)8 (17.8%)
  Red blood cell transfusion2
  Pulmonary edema1
  Adrenal hematoma1
 Grade 2, N (%)3 (6.7%)
  Infection (Herpes zoster)1
  Biloma requiring aspiration1
  Pleural effusion1
Follow-up period (days)*1,245 (840–2,026)

Right-lobe grafts were harvested from all single donors. Ten (22%) donors provided liver grafts for five recipients of dual-graft transplantation. Median postoperative intensive care unit stay was 2 days (range, 1–3 days) and median total hospital stay, including pretransplantation work-up for donors was 14 days (range, 9–24 days).

None of the 72 evaluated living donor candidates experienced complications associated with percutaneous preoperative liver biopsy. Overall donor complication rates, defined by the Clavien classification13 as any alteration from the ideal postoperative course to complete recovery, were estimated at 24.4% (n = 11): eight donors (73%) had grade 1 (minor) and three (27%) had grade 2 (no lasting disability) complications. Grade 1 complications included ileus (n = 4), red blood cell transfusion of less than 3 units (n = 2), pulmonary edema (n = 1), and adrenal hematoma (n = 1). All of these complications improved spontaneously or with conservative management. Grade 2 complications included infection (n = 1; herpes zoster), biloma requiring USN-guided aspiration (n = 1), and right pleural effusion requiring percutaneous pigtail insertion (n = 1). Donors were followed for a median of 1245 days (range, 840–2026 days). At the time of last follow-up, all donors remained alive and well with normal liver function.


We have shown here that ALF in about 90% of adult patients in Korea, an HBV-endemic area, was caused by etiologies associated with low spontaneous recovery rates, including HBV, use of herbal medications, ingestion of drugs other than APAP, AIH, and mushroom poisoning. The high prevalence of these etiologies may have contributed to the poor transplantation-free survival rate in our patients. However, only 4% of patients listed for LT were able to receive liver grafts from deceased donors, whereas about 40% underwent adult LDLT, with a 1-year survival rate of 85%. The present study is unique because it prospectively and integratively analyzed the etiologies of ALF and the effect of adult LDLT in an inception cohort of patients from the time of diagnosis. Although several previous studies have suggested that adult LDLT improves the survival of patients with ALF, most have been small, retrospective cohort reports on patients at the time of transplantation.7–9

It is well known that the etiology of ALF varies considerably by geographical distribution, and is a key factor determining patient outcome. For example, APAP, which is usually associated with a favorable outcome, is the most common cause of ALF in the United States and the United Kingdom.1, 2 By contrast, etiologies with poor outcome, including HBV, are the main causes of ALF in Asia and certain parts of Europe,1, 2 and it would be in these areas that the availability of emergency LT would have the greatest impact on survival. However, the organ supply from deceased donors is extremely limited in most Asian countries.6 The number of deceased donors per 1 million populations is usually less than five in most Asian countries, whereas it ranges between 10 and 35 in Western countries. In Korea, the donation rate is especially low, with fewer than two donors per million inhabitants during our study period.6 Thus, as shown here, few patients are able to undergo DDLT, despite the fact that ALF is a condition with the most urgent transplantation status (KONOS status 1), and the nationwide sharing of organs from deceased donors. Given the rapid progression of ALF and the extreme shortage of livers from deceased donors, adult LDLT would be a promising treatment option in such critically ill patients.

An acceptable recipient survival rate and a guarantee of donor safety are prerequisites that ethically justify the risks taken by adult LDLT donors.14 By multivariate analyses, after adjusting for variables such as age and MELD we found that both adult LDLT and DDLT greatly reduced the 1-year mortality rate of patients when compared with patients who did not undergo LT, with HRs of 0.10 and 0.12, respectively. Our finding that the 1-year survival rate of adult LDLT recipients was 85% was similar to the 1-year survival rate of 82% in patients who underwent DDLT for ALF in the United States3, 15 and to worldwide data on the effect of adult LDLT in ALF patients (70% to 87.5% survival).7–9

The short time from diagnosis to death (median, 7 days) among patients in the no-LT group who died waiting for a graft highlights the limited window during which transplantation is possible. Interestingly, we found that all significant factors predicting 1-year posttransplantation mortality were associated with renal impairment or metabolic derangement. These included dialysis, creatinine concentration, arterial pH, and lactate concentration measured just prior to LT. These results indicate that, for patients with ALF, delayed transplantation may be associated with poor posttransplantation survival and indicate the importance of expediting emergency LT before deterioration of renal function and metabolic status. In this context, adult LDLT would offer an advantage over DDLT, by reducing waiting time and providing more optimal timing of surgery. In the present study the median waiting time from diagnosis to LT was 2.5 days for adult LDLT and 5.5 days for DDLT. Although this difference was not significant, probably because of the small numbers of patients who underwent DDLT, our findings are consistent with previous reports showing that adult LDLT was associated with shorter waiting times.7, 9, 16

In the present study, some of our patients received dual-graft LDLTs, using the same surgical techniques previously described.17 However, we do not advocate the routine use of dual-graft transplantation for patients with ALF because of the potential for an increased risk to donors. Dual-graft transplantation was considered as a last resort when single-graft transplantation did not appear feasible after considering donor safety (remnant volume <30% of total liver volume and/or severe steatosis) and small-for-size graft for recipient.

The patients in the LT group were significantly younger than those in the no-LT group, suggesting that younger patients were more likely to receive LT than were older patients and that this may have contributed, at least in part, to the higher likelihood of death observed in older patients. Multivariate analysis showed, however, that age was a prognostic factor for 1-year mortality independent of LT and MELD. The importance of age as a prognostic factor for ALF has also been reported.3, 18, 19 Although the exact mechanism has not been clarified, the residual capacity for liver regeneration in older patients may be less than in younger patients, or older patients may be at greater risk for complications such as infections and multiple organ failure.

There may be a concern that the expedited donor evaluation could be associated with poorer donor outcomes. However, there was no mortality, major morbidity, or reoperation among the donors of our series. Although about 24% of donors suffered from minor complications, all improved spontaneously or with conservative management. The rate of donor complications we observed was lower than that of adult LDLT donors in a multicenter study performed in the United States (38%),20 but higher than that in large LT centers in Asia (12%–16%).21, 22 The differences may be attributable to variation in defining and reporting complications. In general, however, emergency adult LDLT is not likely to be associated with a significantly higher donor complication rate than is elective adult LDLT.

Another major ethical consideration has been coercion of potential donors. As indicated previously,14, 23 an element of coercion can always exist between any potential donor and recipient. A complete absence of coercion when making a decision to donate may be unrealistic, because of the dynamics of the life-threatening condition of the recipient combined with the life-rescuing possibility of LDLT and the familial relationship between donor and recipient. However, we made maximal efforts to guarantee donor autonomy during acquisition of written informed consent. Freedom of withdrawal was allowed at any stage of the donor evaluation process. A recent systematic review focused on this issue found that nearly all adult LDLT donors reported no coercion to donate,24 and more than 85% of donors reported that the information available to them prior to undergoing the procedure was adequate. Although our institution has favored the use of voluntary donors unrelated to recipients, most living donors in this study were family members of recipients. The reasons for this included the limited number of voluntary living donors, the severely limited time in which LT is available for patients with ALF, and the difficulty in guaranteeing the absence of any form of trade between donor and recipient.

This study had several limitations. First, it was a single-center study and therefore may not represent the entire ALF patient population in HBV-endemic areas. However, our institution is the largest LT center in Korea and our liver transplant program performs nearly half of all yearly transplants in the country, suggesting that our results may reflect the situation throughout our country.25 Second, it was difficult for us to directly compare patient outcomes by etiology, because of the small numbers of patients with ALF etiologies associated with favorable outcomes. Third, the role of emergency adult LDLT for patients with ALF was not analyzed in a randomized controlled manner. However, because of the high rate of mortality without transplantation and the extremely limited availability of DDLT, allocating patients with ALF to a no-LT or DDLT control group would be unethical or impossible.

In conclusion, we have shown here that emergency adult LDLT can be performed expeditiously and safely for patients with ALF, and that the procedure greatly improves patient survival rate. Adult LDLT should therefore be considered one of the first-line treatment options for patients with ALF, especially in areas where most cases of ALF are caused by etiologies associated with poor outcome and the supply of organs from deceased donors is severely limited.


The authors thank Drs. Ki-Hun Kim, Chul-Soo Ahn, Duk-Bog Moon, Tae-Yong Ha, Gi-Won Song, Dong-Hwan Jung, Kang Mo Kim, Young-Hwa Chung, and Yung Sang Lee for their help in data collection and critical review of the article.