Abstract
BACKGROUND
The long-term clinical outcomes in initial survivors with acute liver failure (ALF) are not well known.
AIMS
The aim of the current study is to provide an overview of the 2 year clinical outcomes amongst initial survivors and liver transplant (LT) recipients that were alive 3 weeks after enrollment in the Acute Liver Failure Study Group (ALFSG).
METHODS
Outcomes in adult ALFSG patients that were enrolled between 1998 and 2010 were reviewed.
RESULTS
2-year patient survival was significantly higher in the 262 LT recipients (92.4%) compared to the 306 acetaminophen (APAP) spontaneous survivors (SS) (89.5%) and 200 non-APAP SS (75.5%) (p < 0.0001). The causes of death were similar in the 3 groups but the time to death was significantly longer in the LT recipients (p< 0.0001). Independent predictors of 2-year mortality in the APAP group were a high serum phosphate level and patient age (c-statistic = 0.65 (0.54, 0.76)), patient age and days from jaundice to ALF onset in the non-APAP group (c-statistic =0.69 (0.60, 0.78)), and patient age, days from jaundice, and higher coma grade in the LT recipients (c-statistic=0.74 (0.61, 0.87)). The LT recipients were significantly more likely to be employed and have a higher educational level (p < 0.05).
CONCLUSIONS
Two-year outcomes in initial survivors of ALF are generally good but non-APAP patients have a significantly lower survival which may relate to pre-existing medical co-morbidities. Spontaneous survivors with APAP overdose experience substantial morbidity during follow-up from ongoing psychiatric and substance abuse issues.
Keywords: cerebral edema, liver transplantation, prognosis, regeneration, age
Introduction
Acute liver failure (ALF) is a rare but dramatic illness associated with a high rate of short-term morbidity and mortality (1,2). The likelihood of an early adverse outcome including the need for emergency liver transplantation (LT) is, in part, related to the etiology of ALF, peak grade of encephalopathy, and development of multi-organ failure (1–4). Results of several studies have demonstrated a lower 1-year patient survival in ALF LT recipients compared to those with cirrhosis undergoing LT due to the emergency nature of the surgery, cerebral edema, and frequent need for life support in the ALF patients (2, 3). In addition, a higher rate of allograft dysfunction including acute and chronic rejection, de novo autoimmune hepatitis, and recurrent graft failure have been demonstrated during follow-up in ALF LT recipients (5–7). Furthermore, variable degrees of neurological sequelae in long-term survivors of ALF have been reported but the number of patients evaluated has been small (8,9).
The US Acute Liver Failure Study Group (ALFSG) is a consortium of 13 academic medical centers that is conducting an ongoing observational registry study of adult ALF patients (2). Important insights into the etiology, disease specific prognosis, and clinical correlates of early adverse outcomes have been reported (1,2). In addition to collecting detailed clinical, demographic, and outcome data up to 3 weeks after enrollment, the study includes a prospective assessment of patient outcomes at 1 and 2 years after enrollment. This effort has proven useful in better defining the long-term clinical outcomes in specific subgroups of ALF patients (10,11). The aim of the current study is to provide an overview on the 2 year clinical outcomes amongst initial survivors and LT recipients that were alive 3 weeks after enrollment in the ALFSG. Baseline demographics and clinical factors associated with impaired long-term survival were identified as well as the frequency of various clinical complications amongst the following three groups: acetaminophen (APAP) initial spontaneous survivors (SS), non-APAP initial SS, and LT recipients. We hypothesized that LT recipients would have poorer long-term outcomes due to their more severe illness necessitating emergency surgery. Multivariate models to identify patients at risk of dying within 2 years of follow-up were constructed to provide prognostic data for clinicians.
Patients and Methods
Patient Selection
The ALFSG database consists of 1850 adult patients with ALF admitted to one of 13 actively or 10 previously enrolling clinical sites from January 1, 1998 to July 1st, 2010. Eligible ALF patients had symptoms of jaundice or illness of less than 26 weeks prior to admission and mental status changes with coagulopathy defined as an international normalized ratio (INR) > 1.5 without known underlying chronic liver disease. The protocol was reviewed and approved by the Institutional Review Board at all of the participating sites and written informed consent was obtained from the patient’s next of kin. In addition to baseline demographics and presenting features, detailed information during the first seven days of enrollment including imaging, lab, and clinical data, as well as short term outcome data were obtained through 21 days after admission.
Long-term follow-up protocol and data collection
In calendar year 2000, the ALFSG observational study was amended to include a 12 month study visit to be performed at 6 to 18 months after enrollment as well as a 24 month study visit completed any time after 18 months from enrollment. Data obtained at these visits included incremental medical history, medications, liver biopsy and imaging data, and laboratory variables. In addition various complications were prospectively tracked including neurological complications. Data were collected at a face to face study visit or via telephone interview and/or chart review. For patients that were lost to follow-up, sites queried the Social Security death index database and reviewed medical records. To be included in the analysis population, patients must have survived to short-term follow-up at 21 days (1,138 patients) and either have a recorded death or at least one long term follow-up form completed 21 days after enrollment (773 patients). For the current analysis, patients were stratified on whether they received a liver transplant (LT), and the 3-week initial spontaneous survivors were further stratified by presence of APAP- related or other non-APAP etiology.
Statistical Analysis
SAS software version 9.2 (SAS Institute Inc. Cary, NC) was used to perform statistical analysis. Baseline variables were described using counts and percentages for categorical data, or means and standard deviations (medians and interquartile ranges) for continuous normal (skewed) data. For variables identified as clinically relevant, statistical tests were performed using chi-square, ANOVA, or Kruskal-Wallis tests. Due to the low event rate in each of the populations, 2-year mortality was reported using cumulative percent and tested using a test of independent proportions. Predictors of death were modeled with univariate and multivariate logistic regression, with predictors identified using step-wise elimination based on significance and overall goodness of fit. Variables with a p-value <0.20 on univariate analyses were considered for the multivariable analysis Results of the modeling are expressed as odds ratios with 95% confidence intervals. The c-statistic and corresponding 95% confidence interval are based on an approximation (12).
RESULTS
There were 1,138 adult ALF patients alive at 3 weeks after initial enrollment into the ALFSG registry between January 1998 and July 2010 (Figure 1). Long-term follow-up data were available in 773 (68%) of these patients. Amongst the 773 patients included in this analysis, there were 262 LT recipients and 511 initial spontaneous survivors (SS) which included 306 with APAP overdose and 200 with ALF due to other non-APAP etiologies (Table 1). The patients not included in this analysis were similar in baseline demographic and clinical features except that the excluded population was more likely to have experienced APAP overdose (Supplemental Table 1).
Figure 1. ALF patients included in the current analysis.
Amongst the 773 ALF patients eligible for this study, there were 262 liver transplant (LT) recipients and 511 initial Spontaneous survivors (SS) that were alive 3 weeks after enrollment into the ALFSG registry.
Table 1.
Baseline characteristics of the long-term follow-up ALF patient groups
| N | APAP Initial Survivors N=306 |
Non-APAP Initial Survivors N=200 |
LT recipient N=262 |
P-value | ||
|---|---|---|---|---|---|---|
| Demographics | ||||||
| Age (yrs) | 768 | 37.5±12.5 | 42.6±15.2 | 38.9±13.7 | < 0.001 | |
| % Female | 768 | 235(76.8) | 122(61.0) | 167(63.7) | < 0.001 | |
| Race | % Caucasian | 768 | 266(86.9) | 138(69.0) | 183(69.8) | < 0.001 |
| % Black | 23(7.5) | 41(20.5) | 46(17.6) | |||
| % Other | 17(5.6) | 21(10.5) | 33(12.6) | |||
| Ethnicity (% Not Hispanic/Latino) | 768 | 293(95.8) | 182(91.0) | 231(88.2) | 0.004 | |
| Marital Status | % Married | 495 | 78(39.6) | 58(52.3) | 108(57.8) | 0.01 |
| % Never Married | 79(40.1) | 37(33.3) | 54(28.9) | |||
| % Other | 40(20.3) | 16(14.4) | 25(13.4) | |||
| Employment | % Employed | 768 | 67(21.9) | 50(25.0) | 114(43.5) | < 0.001 |
| % Other | 44(14.4) | 33(16.5) | 43(16.4) | |||
| % Unknown | 195(63.7) | 117(58.5) | 105(40.1) | |||
| Years of education | 474 | 12.5±2.4 | 12.9±2.6 | 13.5±2.6 | 0.001 | |
| Medical history | ||||||
| Etiology | % APAP | 764 | 306(100.0) | 0(0.0) | 33(12.8) | < 0.001* |
| % Autoimmune Hepatitis | 0(0.0) | 21(10.5) | 37(14.3) | |||
| % DILI | 0(0.0) | 48(24.0) | 55(21.3) | |||
| % Hepatitis A | 0(0.0) | 13(6.5) | 7(2.7) | |||
| % Hepatitis B | 0(0.0) | 22(11.0) | 32(12.4) | |||
| % Shock/Ischemia | 0(0.0) | 38(19.0) | 1(0.4) | |||
| % Indeterminate | 0(0.0) | 33(16.5) | 68(26.4) | |||
| % Other | 0(0.0) | 25(12.5) | 25(9.7) | |||
| Symptoms to ALF (days) | 739 | 3.0±2.0 | 10.0±14.0 | 17.0±23.2 | <0.001 | |
| Jaundice to ALF (days) | 635 | 1.0±2.0 | 5.0±8.5 | 11.5±17.0 | <0.001 | |
| Weight(kg) | 728 | 72.5±19.8 | 80.6±21.3 | 80.2±20.7 | <0.001 | |
| % Psychiatric disease | 768 | 167(54.6) | 54(27.0) | 34(13.0) | < 0.001 | |
| % Substance abuse | 768 | 140(45.8) | 29(14.5) | 21(8.0) | < 0.001 | |
| % History of IDU | 756 | 41(13.6) | 16(8.1) | 4(1.6) | < 0.001 | |
| % Hypertension | 768 | 32(10.5) | 53(26.5) | 34(13.0) | < 0.001 | |
| % Endocrine/Diabetes | 768 | 31(10.1) | 40(20.0) | 34(13.0) | 0.006 | |
| % Heart disease | 768 | 14(4.6) | 25(12.5) | 11(4.2) | < 0.001 | |
| Presenting labs | ||||||
| AST (IU/L) | 765 | 2805±6300 | 1048±2070 | 626±1431 | <0.001 | |
| ALT (IU/L) | 761 | 3272±4226 | 1454±2358 | 766±1762 | <0.001 | |
| INR | 759 | 2.5±2.0 | 2.1±1.2 | 3.1±2.3 | <0.001 | |
| Bilirubin (mg/dL) | 764 | 4.1±3.6 | 9.9±14.6 | 22.1±16.3 | <0.001 | |
| MELD | 754 | 27.8±8.9 | 29.3±7.1 | 34.8±7.4 | <0.001 | |
| Phosphate (IU/L) | 655 | 2.3±1.8 | 3.3±2.1 | 3.3±2.1 | <0.001 | |
| Creatinine (mg/dL) | 767 | 1.4±2.1 | 1.4±1.9 | 1.1±1.6 | 0.055 | |
| Lactate (mmol/L) | 396 | 3.2±3.6 | 3.5±3.3 | 4.9±6.7 | <0.001 | |
| % Urine tox screen positive | 768 | 180(58.8) | 48(24.0) | 48(18.3) | < 0.001 | |
| Clinical complications at admission | ||||||
| % Pressors | 748 | 32(10.6) | 29(14.9) | 37(14.7) | 0.259 | |
| % Intubated | 766 | 130(42.5) | 72(36.2) | 95(36.4) | 0.228 | |
| % Grade ¾ HE | 766 | 134(43.8) | 70(35.2) | 98(37.5) | 0.115 | |
| % Mannitol | 759 | 29(9.6) | 13(6.5) | 37(14.4) | 0.02 | |
| % ICP Monitor | 705 | 21(7.3) | 11(6.0) | 31(13.3) | 0.015 | |
| % Dialysis/CVVH | 753 | 63(20.8) | 28(14.2) | 51(20.2) | 0.150 | |
| Peak Labs | ||||||
| ALT (IU/L) | 768 | 3677±4669 | 1479±2391 | 939±2071 | <0.001 | |
| INR | 763 | 2.8±2.3 | 2.4±1.1 | 3.8±2.9 | <0.001 | |
| Bilirubin (mg/dL) | 768 | 7.0±8.2 | 14.2±16.4 | 24.8±14.0 | <0.001 | |
| MELD | 763 | 29.6±9.2 | 31.0±7.2 | 37.1±7.4 | <0.001 | |
| Creatinine (mg/dL) | 768 | 1.9±3.9 | 1.7±2.8 | 1.4±2.1 | 0.041 | |
| Lactate (mmol/L) | 449 | 3.3±3.9 | 3.7±3.6 | 4.9±6.6 | <0.001 | |
| Clinical Complications during ALF hospitalization | ||||||
| % Pressors | 768 | 57(18.6) | 38(19.0) | 71(27.1) | 0.029 | |
| % Intubated | 768 | 160(52.3) | 85(42.5) | 159(60.7) | < 0.001 | |
| % Grade ¾ HE | 766 | 162(52.9) | 89(44.7) | 149(57.1) | 0.03 | |
| % Mannitol | 768 | 51(16.7) | 19(9.5) | 58(22.1) | 0.001 | |
| % ICP Monitor | 768 | 36(11.8) | 13(6.5) | 60(22.9) | < 0.001 | |
| % Dialysis/CVVH | 768 | 99(32.4) | 52(26.0) | 86(32.8) | 0.222 | |
Compares the proportion of non-APAP etiologies between the non-APAP SS and LT recipients
DILI = Drug induced liver injury IDU= intravenous drug use
The 306 patients in the APAP initial SS group were younger and significantly more likely to be female, Caucasian, and of non-Hispanic ethnicity compared to the other two patient groups. As expected, the APAP initial survivors had the shortest duration of symptoms, the highest serum aminotransferase levels and lowest bilirubin levels at presentation. The APAP initial survivors were also significantly more likely to have active psychiatric and substance abuse co-morbidities at enrollment and a positive urine toxicology screen. A careful review of the cases demonstrated that 116 of these were intentional APAP overdose patients and 159 were non-intentional therapeutic misadventures while 31 were of unknown intent. The majority of LT recipients were also women and they had a significantly longer duration of symptoms to jaundice, higher total bilirubin and lower serum aminotransferase levels at presentation compared to the other two groups. The frequency of observed etiologies of ALF in the LT recipients differed when compared to non-APAP initial SS (p-value ≤ 0.001). The LT recipients also had the highest INR, lactate, and model for end-stage liver disease (MELD) scores at presentation. Although the level of hepatic encephalopathy was similar in all 3 patient groups at admission, the LT recipients were more likely to have an intracranial pressure (ICP) monitor placed, require mechanical intubation, and receive pressors and mannitol during their illness compared to the other two patient groups.
2-year patient survival
The 2-year actuarial patient survival was highest in the LT recipients (92.4%) followed by the APAP initial SS (89.5%) and non-APAP initial SS (75.5%). Cumulative death rates are presented in Figure 2. The median time to death in the LT recipients of 223 days (range: 23 to 568 days) was significantly longer compared to the other two patient groups (p-value <0.001). Overall, the leading causes of death were liver related (17%), infection/sepsis (12%), and multisystem organ failure (18%) but 45% of the deaths were of unknown cause (Table 2). The identified causes of death were similar in the 3 patients groups.
Figure 2. Actuarial 2-year mortality amongst the 773 initial 3-week survivors.
The liver transplant recipients (C) had the highest 2-year survival (92.4%) compared to the Acetaminophen overdose (A) spontaneous survivors (89.5%) and the non-Acetaminophen (B) spontaneous survivors (75.0%) (p < 0.0001).
Table 2.
Causes of death in ALF patients
| APAP Initial Survivors N=32 |
Non-APAP Initial Survivors N=49 |
LT recipients N=20 |
|
|---|---|---|---|
| % Liver related | 9.4 | 18.4 | 25.0 |
| % Infection/sepsis | 12.5 | 12.2 | 10.0 |
| % Cardiac | 0.0 | 8.2 | 10.0 |
| % Neurological | 6.3 | 4.1 | 10.0 |
| % Multisystem organ failure | 28.1 | 8.2 | 25.0 |
| % Other | 12.5 | 6.1 | 15.0 |
| % Unknown | 40.6 | 53.1 | 30.0 |
| Days to death Median (IQR)* | 57.0 (84.0) | 57.0 (95.0) | 222.5 (380.5) |
P-Value: <0.001 Patients can have multiple causes of death
Predictors of impaired long-term survival
Multivariate models were constructed to identify baseline variables associated with impaired 2-year survival in each of the 3 patient groups (Table 3). Univariate analyses of the admission clinical data were used to identify potential variables for model building.
Table 3.
Univariate and multivariate models of 2-year survival in the 3 ALF patient groups
| Univariate Odds Ratio (95% CI) | P-Value | Multivariate Odds Ratio (95% CI) | P-Value | |
|---|---|---|---|---|
| APAP Initial Spontaneous Survivors | C-Statistic: 0.647 (0.539,0.755) | |||
| Age (per 10 yrs) | 1.34 (1.02,1.76) | 0.038 | 1.27 (0.95, 1.71) | 0.110 |
| ALT (per 50 IU/L) | 0.994 (0.987, 1.001) | 0.071 | --- | --- |
| Phosphate (mg/dL) | 1.28 (1.07,1.54) | 0.007 | 1.26 (1.05,1.51) | 0.015 |
| Arterial pH | 0.04 (0.001,1.125) | 0.059 | --- | --- |
| Initial HE (%3/4) | 2.02 (0.96,4.26) | 0.064 | --- | --- |
| Pressor Use | 2.73 (1.07,6.95) | 0.035 | --- | --- |
| Dialysis/CVVH Use | 1.87(0.84,4.18) | 0.128 | --- | --- |
| Non-APAP Initial Spontaneous Survivors | C-Statistic: 0.687 (0.596,0.778) | |||
| Age (per 10 yrs) | 1.44 (1.15,1.80) | 0.001 | 1.54 (1.17,2.02) | 0.002 |
| Ethnicity (%Hisp/Lat) | 0.36 (0.08,1.62) | 0.183 | --- | --- |
| Education (% <12 yrs) | 2.16 (0.76,6.10) | 0.148 | --- | --- |
| Days from Jaundice to ALF | 1.03 (1.01,1.06) | 0.015 | 1.03 (1.00,1.06) | 0.028 |
| ALT (per 50 IU/L) | 0.987 (0.978, 0.997) | 0.013 | --- | --- |
| INR | 0.83 (0.64,1.06) | 0.138 | --- | --- |
| Pressor Use | 1.77 (0.76,4.14) | 0.185 | --- | --- |
| Liver transplant recipients | C-Statistic: 0.741 (0.613,0.869) | |||
| Age (10 yrs) | 1.43 (1.02,2.01) | 0.036 | 1.53 (1.06,2.21) | 0.024 |
| Days from Jaundice to ALF | 0.97 (0.93,1.01) | 0.113 | 0.97 (0.93,1.01) | 0.162 |
| Days from Symptoms to ALF | 0.98 (0.95,1.01) | 0.18 | --- | --- |
| Initial HE (%3/4) | 2.70 (1.06,6.87) | 0.037 | 2.57 (0.94, 7.05) | 0.066 |
| INR | 0.78 (0.58,1.05) | 0.104 | --- | --- |
| MELD | 0.96 (0.90,1.02) | 0.173 | --- | --- |
| Intubation Use | 2.28 (0.91,5.73) | 0.078 | --- | --- |
| Dialysis/CVVH Use | 2.31 (0.87,6.14) | 0.092 | --- | --- |
Odds are modeled as the probability of death
Acetaminophen Initial Spontaneous Survivors
Older patient age, serum alanine aminotransferase (ALT) as well as elevated serum phosphate, lower arterial pH, and higher initial coma grade were significant predictors of death on univariate analysis. In the multivariate model, age and phosphate remained as consistent predictors of death with an elevated phosphate level indicating higher risk of death (OR: 1.26, 95% CI: 1.05, 1.51). The median age of the APAP patients that died during follow-up was 42 while the median age of the long-term survivors was 35. The median admission phosphate of those that died was 2.75 mg/dl compared to 2.20 mg/dl in those that survived. Of note, the proportion of patients with intentional APAP overdose was similar in the long-term survivors and non-survivors (41% vs 42%, p= 0.93) as was the presence of psychiatric co-morbidities (44% vs 56%, p= 0.19).
Non-Acetaminophen Spontaneous Survivor group
Older patient age, non-Hispanic ethnicity, <12 years of education, increased days of jaundice to ALF, serum ALT, higher INR and pressor use were examined as potential predictors of death for the multivariate analysis. Of note, initial coma grade, need for dialysis, and intubation were not associated with long-term survival in univariate analyses. In the multivariate model, age and days from jaundice to ALF were the only independent predictors of death in this patient group. The median age of the non-APAP patients who died was 50 versus 42 in those that survived. The etiologies of ALF in those who died varied but there was no specific non-APAP etiology which was over-represented in those who died versus survived (Supplemental Table 2a, p=0.35). Of the 49 patients who died with non-APAP ALF, only 5 (10.2%) were listed for liver transplant while 32 of the 151 survivors (21.1%) were initially listed for LT during their hospitalization.
Liver transplant recipients
In the multivariate model for LT recipients, age, days from jaundice to ALF onset, and higher coma grade remained as consistent predictors of death with older patient age indicating higher risk of death (ten yr OR: 1.53, 95%CI: 1.06,2.21). There did not appear to be an over-representation of any particular etiology of ALF in those who died versus survived (Supplemental Table 2b, p=0.07). However, 12% of the APAP LT recipients died as well as 11% of the autoimmune patients and 5% of the drug induced liver injury (DILI) patients. Although the proportion of patients with advanced encephalopathy (Grade ¾) at presentation was significant on univariate analysis, this did not persist in multivariate analysis. The median age of the LT recipients who died was 45.5 compared to 38 years in those who survived. The median days to jaundice of the LT recipients who died was 6 compared to 12 in those who survived. The proportion of LT recipients who died that had grade ¾ HE was 60% compared to 36% in those who survived.
Long-term clinical and functional outcomes
Assessment of long-term functional outcomes are reported in Table 4 from the 353 patients who provided follow-up data in person or via telephone interview including 123 APAP initial survivors, 85 non-APAP initial survivors, and 145 LT recipients. The rate of employment, marriage, and having private insurance was lowest in the APAP patients compared to others. In the subgroup of patients with an available Karnofsky score, the median score among the APAP initial survivors was lower than the other two groups but this difference was not statistically significant.
Table 4.
Functional status at last follow-up visit among ALF long-term survivors
| N * | APAP Initial Survivors (N=123) |
Non-APAP Initial Survivors (N=85) |
LT Recipients (N=145) |
P-Value | ||
|---|---|---|---|---|---|---|
| Demographics | ||||||
| Time of Form Collection | 276 | 704±435 | 732±424 | 746±400 | 0.191 | |
| Age (years) | 353 | 37.5±12.8 | 42.2±13.9 | 39.2±13.0 | 0.044 | |
| % Female | 353 | 95(77.2) | 53(62.4) | 95(65.5) | 0.040 | |
| Marital Status | % Married | 247 | 24(30.8) | 33(54.1) | 56(51.9) | 0.025 |
| % Never Married | 247 | 34(43.6) | 19(31.1) | 30(27.8) | ||
| % Other | 247 | 20(25.6) | 9(14.8) | 22(20.4) | ||
| Employment | % Employed | 276 | 27(29.0) | 30(45.5) | 53(45.3) | < 0.001 |
| % Other | 276 | 35(37.6) | 26(39.4) | 52(44.4) | ||
| % Unknown | 276 | 31(33.3) | 10(15.2) | 12(10.3) | ||
| Years of Education | 197 | 11.8±2.3 | 12.7±2.3 | 12.6±3.2 | 0.181 | |
| % Uninsured | 276 | 14(15.1) | 5(7.6) | 2(1.7) | 0.001 | |
| % Private Insurance | 276 | 36(38.7) | 45(68.2) | 73(62.4) | < 0.001 | |
| % Medicare | 276 | 16(17.2) | 9(13.6) | 17(14.5) | 0.796 | |
| % Medicaid | 276 | 12(12.9) | 5(7.6) | 18(15.4) | 0.312 | |
| Karnofsky Score | 197 | 86±16 | 90±12 | 90±11 | 0.148 | |
| Substance abuse at last follow-up visit | ||||||
| Time of Form Collection | 344 | 762±408 | 734±443 | 746±346 | 0.38 | |
| % Suicide attempt | No | 325 | 80(72.7) | 59(76.6) | 124(89.9) | 0.004 |
| Yes | 325 | 7(6.4) | 2(2.6) | 3(2.2) | ||
| Unknown | 325 | 23(20.9) | 16(20.8) | 11(8.0) | ||
| % Psychiatric disease | No | 326 | 40(36.0) | 48(62.3) | 96(69.6) | < 0.001 |
| Yes | 326 | 45(40.5) | 14(18.2) | 31(22.5) | ||
| Unknown | 326 | 26(23.4) | 15(19.5) | 11(8.0) | ||
| % Alcohol Abuse | No | 325 | 71(64.0) | 55(72.4) | 126(91.3) | < 0.001 |
| Yes | 325 | 14(12.6) | 4(5.3) | 1(0.7) | ||
| Unknown | 325 | 26(23.4) | 17(22.4) | 11(8.0) | ||
| Medical complications at last follow-up visit | ||||||
| Time of Form Collection | 353 | 763±375 | 738±444 | 746±370 | 0.601 | |
| % Neurological Complications | 353 | 33 (39.3) | 16 (26.7) | 42 (32.6) | 0.276 | |
| % Diabetes Mellitus | 270 | 8(9.8) | 8(13.3) | 22(17.2) | 0.314 | |
| % Hypertension | 256 | 15(18.8) | 13(22.0) | 27(23.1) | 0.763 | |
| % Hyperlipidemia | 267 | 9(11.4) | 7(11.7) | 9(7.0) | 0.454 | |
| Most recent labs | ||||||
| Time of Last Lab | 320 | 224 ±672 | 264±695 | 709±449 | <0.001 | |
| Median AST (IU/L) | 286 | 27±46 | 30±34 | 31±27 | 0.669 | |
| % AST > ULN | 286 | 41(45.6) | 26(42.6) | 69(51.1) | 0.491 | |
| Median ALT (IU/L) | 287 | 39±99 | 28±44 | 31±33 | 0.613 | |
| % ALT > ULN | 287 | 47(52.2) | 26(41.9) | 58(43.0) | 0.316 | |
| Median Alk phos (IU/L) | 226 | 82±33 | 82±69 | 89±78 | 0.336 | |
| % Alk phos > ULN | 226 | 13(20.3) | 14(31.1) | 33(28.2) | 0.383 | |
| Median T. bilirubin (mg/dL) | 284 | 0.7±1.3 | 0.8±2.0 | 0.6±0.4 | 0.015 | |
| % bilirubin > ULN | 284 | 30(33.7) | 24(40.0) | 22(16.3) | <0.001 | |
| Median creatinine (mg/dL) | 295 | 0.8±0.6 | 0.9±0.5 | 1.1±0.5 | <0.001 | |
| % creatinine > ULN | 295 | 20(21.3) | 9(14.3) | 39(28.3) | 0.082 | |
| Median hemoglobin (g/dL) | 291 | 12.1±2.8 | 12.9±3.2 | 12.7±2.1 | 0.033 | |
| % hemoglobin < LLN | 291 | 73(76.8) | 38(64.4) | 96(70.1) | 0.237 | |
| Median platelet count(x 103/ml) | 293 | 248±140 | 210±137 | 212±89 | 0.011 | |
| % platelet count < LLN | 293 | 13(13.8) | 14(23.0) | 22(15.9) | 0.313 | |
Patient population defined by collection of outcome data at in person or via phone with chart review.
The following ULN values were used: AST ULN 30 IU/L, ALT ULN 35 IU/l, alk phos ULN 130 IU/l, bilirubin ULN 1.2 mg/dl, creatinine ULN 1.3 mg/dl. The following LLN were used: Hemoglobin LLN 13.5 g/dl, platelet count LLN 150
Medical and psychiatric co-morbidities during long-term follow-up
The frequency of diabetes, hypertension, and hyperlipidemia in the three groups were similar at the last available long-term follow-up study visit (Table 4). However, ongoing psychiatric illness was significantly more common in the APAP SS patients compared to the other groups. Furthermore, problematic alcohol consumption was significantly more commonly reported in the APAP group compared to the others and intentional drug overdose/suicide was also significantly more common in the APAP SS patients.
Liver disease and laboratory status
The median values of serum AST, ALT, alk phos and bilirubin were normal at the last available follow-up visit in the 3 subgroups of patients. However, the median serum total bilirubin levels were significantly lower in the LT recipients compared to the other patient groups while the median serum creatinine levels were significantly higher in the LT recipients (Table 4).
Discussion
ALF is a rare but potentially devastating illness that frequently afflicts previously healthy and young individuals. The ALFSG and other groups have previously reported that 3- week transplant-free survival is significantly higher in subjects with APAP overdose, hepatitis A, and pregnancy related ALF (~60 to 70%) compared to subjects with DILI, Wilson’s disease and indeterminate ALF (~ 20 to 40%) (1–4). In addition, the ALFSG demonstrated that patient age, initial and peak grade of hepatic encephalopathy, and selected laboratory variables are associated with a greater likelihood of death during the initial ALF hospitalization (1,2). Currently, many clinicians use these parameters and other laboratory and clinical features at the bedside to help guide decisions regarding the need to list and proceed with LT (3,4). However, the long-term clinical outcomes in ALF patients who survive their initial illness are not well described. Understanding long-term outcomes in initial survivors with ALF has proven difficult due to the low incidence of the disease in the general population with only 2,000 to 3,000 annual cases in the US.
In the current study, we were able to determine the 2-year clinical outcomes in 773 ALF patients who had survived their initial illness. The patients included in this analysis are generally representative of the entire ALFSG cohort although fewer APAP patients were included compared to the overall cohort (Supplemental Table 1). Liver transplantation yielded the highest 2-year survival at 92.4% compared to the other 2 groups of initial spontaneous survivors. These results are somewhat surprising since subjects who require an emergency LT tend to be sicker at presentation compared to subjects that spontaneously recover and would be expected to continue to have a substantial risk of death during prolonged follow-up. The greater initial severity of illness at presentation amongst the LT recipients is supported by the higher initial and peak serum bilirubin, INR, lactate and MELD scores in these patients compared to the other two groups (Table 1). The 20 deaths in the LT recipients occurred throughout the 2-years of follow-up while most of the deaths in the initial spontaneous survivors tended to occur within 6 months of enrollment into the ALFSG study, suggesting that ongoing hepatic damage may lead to fatal outcomes after the initial illness. In attempting to understand the reasons for the absolute differences in mortality and time to death, we first looked at the causes of death in the 3 patient groups. There was no apparent differences in the stated causes of death but there were moderate number of “unknown” causes in each subgroup (Table 2). A recent study of hospitalized adult patients with APAP overdose in Canada demonstrated that many of the patients died during 5 years of follow-up from a variety of preventable causes including trauma, substance abuse and other high-risk behaviors (11). In addition, the survival of the APAP overdose patients was significantly lower than the expected rate in age-matched population controls. Therefore, the observation that roughly 10% of our APAP initial survivors died during 2-years of follow-up is not entirely surprising. Our data also suggest that the increased frequency of medical and psychiatric co-morbidities in these high-risk patients may require additional treatments and interventions to improve long-term outcomes.
To further explore the observed differences in mortality, models of 2-year patient mortality were constructed using baseline variables in all 3 patient subgroups (Table 3). Of note, a higher patient age at admission was significantly associated with impaired survival in both the LT and non-APAP initial survivor groups. These data are consistent with results of prior studies of LT recipients which have also demonstrated that increased patient age is a strong negative predictor of survival in ALF patients (14–16). Other studies have also demonstrated that liver regeneration is impaired in older livers (14, 17). Therefore, our data are consistent with prior studies demonstrating the importance of subject age in surviving severe acute liver injury. In each of the subgroups, specific baseline features were identified as being associated with impaired survival on multivariate analysis (Table 3). Amongst the APAP patients, a higher phosphate level was associated with poorer survival which is consistent with prior studies demonstrating that hyperphosphatemia is associated with poorer short-term outcomes (18, 19). This association may be due to impaired utilization of phosphate in liver regeneration or simply as a marker of patients with poorer renal function. Unfortunately, serial serum phosphate levels during the ALF hospitalization or during long-term follow-up were not available for testing. In addition, a lower arterial pH and higher arterial lactate levels are important predictors of short-term outcomes in the King’s college criteria and arterial pH and INR were significant but only on univariate analysis in our cohort (3,4). However, we did not detect a poorer long-term survival in APAP patients with more advanced encephalopathy at presentation. Nonetheless, other studies have demonstrated that severe cerebral edema is predictive of poor neurological outcomes and survival in ALF patients (20,21).
Amongst the non-APAP initial survivors, age, serum ALT levels, and a longer duration of symptoms were all identified on univariate analyses as associated with a lower likelihood of long-term survival (3,23,24). The ALFSG has previously shown that Hispanic patients are more likely to have a lower transplant-free survival rate compared to non-Hispanic patients in part due to their greater likelihood of having DILI (22). However in multivariate analysis, only patient age and a longer duration of symptoms to ALF onset were significant and independent predictors of death. A poorer outcome in patients with subacute ALF that present with more slowly but inevitably progressive liver injury than those with hyperacute ALF has been previously reported (25, 26). Furthermore, the majority of these non-APAP patients were never listed for LT and nearly 20% had ALF due to ischemic hepatitis which is associated with severe underlying cardiopulmonary disease that frequently precludes LT (10).
Amongst the LT recipients, patient age and more advanced encephalopathy at presentation were associated with poorer 2-year survival on univariate analyses. These data are consistent with prior studies that demonstrate the critical importance of the initial and peak encephalopathy grade in predicting outcome in ALF (3, 27, 28). However, in multivariate analysis only age and the days from jaundice to ALF onset remained independent predictors of outcomes. Although other studies have demonstrated that LT recipients with APAP overdose are at increased risk of graft failure compared to other ALF patients, we were not able to identify a relationship between the etiology of ALF and death during prolonged follow-up possibly due to the limited number of deaths in this population (14).
A careful review of the functional outcomes in the 3 patient groups demonstrated some important and interesting differences (Table 4). For example, LT recipients were significantly more likely to be employed compared to the other two groups both at presentation and during follow-up. These observations may in part be due to the lower socioeconomic status of patients who present with APAP overdose compared to the other two groups. Although nearly 40% of the APAP overdose patients were non-intentional “therapeutic misadventures” who received toxic doses of multiple APAP products there was a significantly higher rate of concomitant substance abuse and psychiatric disease requiring treatment in the APAP SS group during follow-up compared to the other two groups which was independent of intentionality (29). In addition, 6.1% of the APAP initial survivors reported another episode of drug overdose/suicide which is also consistent with prior studies of repeated high risk behaviors in this patient population (13).
Recent studies have demonstrated that LT recipients with ALF have a lower rate of employment compared to cirrhotic LT recipients despite their younger age (30). In addition, ALF patients experience higher rates of depression, anxiety, and post-traumatic stress disorder after liver transplantation compared to cirrhotic LT recipients (31). Whether this is due to the sudden onset of a severe life-threatening illness in generally young and previously well ALF patients compared to chronically ill cirrhotic LT recipients is not known but plausible. The Karnofsky scores which are indicative of global health status tended to be higher in the LT recipients compared to the other two groups although the difference did not reach statistical significance. In a recent analysis of health related quality of life measures in a subgroup of ALFSG long-term survivors, we showed that the LT recipients had significantly higher SF-36 and CDC-14 scores during follow-up compared to the APAP and non-APAP initial survivors (31). These cumulative data suggest that both APAP and non-APAP initial survivors appear to have poorer general physical and mental health compared to the LT recipients during long-term follow-up in addition to a lower overall survival. As such, a significant subset of ALF spontaneous survivors may benefit from targeted interventions including counseling, psychiatric care, and physical therapy to achieve improved long-term health outcomes.
The median serum AST, ALT, alk phos, and bilirubin levels were normal in all of the patients in each of the 3 patient subgroups during their most recent follow-up visit (Table 4). The significantly lower total bilirubin levels in the LT recipients is somewhat surprising and suggests that the APAP and non-APAP initial survivors may have poorer bilirubin clearance. However, our study was not designed to assess liver disease status via histological sampling or radiological imaging to further explore this observation. We surmise that the higher mean serum creatinine levels in the LT recipients likely relates to the use of nephrotoxic calcineurin inhibitors to prevent rejection. However, only 28% of the LT recipients had an elevated serum creatinine level so the degree of renal insufficiency appears to be mild. In addition, the incidence of diabetes and hypertension was generally low and similar in the 3 patient groups.
Limitations of the current study include the incomplete follow-up of all 1172 eligible ALF survivors at 3 weeks (Figure 1). One of the reasons for incomplete follow-up is that over 80 % of the ALF patients are transferred to an ALFSG site for tertiary liver care. For patients that survive their index hospitalization with no evidence of residual liver injury, continued follow-up at a referral center that may be several hours from their home, can be challenging. This may in part explain the larger proportion of APAP initial survivors who did not return for follow-up (Supplemental Table 1). In addition, continued participation in a non-interventional registry study would be expected to have a lower rate of follow-up than a clinical trial. Lastly, 45% of the ALF patients that died during prolonged follow-up did not have an identifiable cause of death (Table 2). This rate of incomplete data is similar to those seen in other population and cohort based studies of long-term outcomes of patients (32,33). The incomplete reporting of the causes of death in our study may again relate to the absence of long-term medical follow-up in the liver center due to logistical issues as well as our inability to routinely obtain Death Certificates or causes of death from the Social Security Death Index. In addition, the declining use of autopsy in patients that die in a hospital setting and the difficulty to accurately ascertain the cause of death in individuals without an autopsy may have played a role (32,34,35). Despite these limitations, our study represents the largest cohort of adult ALF patients that were prospectively followed beyond 21 days and up to 2 years. In addition, the severity of illness and other clinical features of the enrolled patients were generally similar to the excluded patients. Another limitation of the current study was that we did not account for donor, operative, or immunosuppression variables when attempting to explore outcomes in the LT recipients. This was due to the fact that our study was not intended to standardize the treatment or monitoring of ALF patients after LT. Furthermore, the limited number of deaths in the LT recipients may have precluded adequate power to identify other predictors of long-term outcomes.
In conclusion, our study results demonstrate that the 2-year clinical outcomes in early survivors with ALF are generally very good. Although nearly 40% of initial adult ALF patients die during their index hospitalization, the mortality rate among initial 3-week survivors is substantially lower during more prolonged follow-up. Contrary to our initial hypothesis, the non-APAP initial survivors have a significantly lower 2-year survival compared to the LT recipients and APAP initial survivors. Most of the deaths in the initial spontaneous survivors occurred within the first 6 months of observation due to a multitude of causes suggesting that early careful follow-up after an ALF episode is warranted. Amongst the long-term survivors, the APAP initial survivors had the lowest rates of employment which may be due to their lower baseline socioeconomic status and more frequent medical and psychiatric co-morbidities compared to the other patient groups. Laboratory markers of liver injury demonstrated normal or near normal parameters in most patients during follow-up. Therefore, despite the need for intubation, dialysis, and pressors in many ALF patients during their acute illness, most will recover with intact liver function during follow-up. Although mortality is highest in the non-APAP initial survivors, the overall favorable 2 year survival can help guide future studies of additional interventions to improve clinical outcomes in these patients.
Supplementary Material
Acknowledgments
Dr. Rangnekar is supported by the T32 DK62708-01, NIDDK Training Grant in Gastrointestinal Epidemiology, and a Clinical and Translational Science Award from the Michigan Institute for Clinical and Health Research.
Members and institutions participating in the Acute Liver Failure Study Group 1998–2010 are as follows: W.M. Lee, M.D. (Principal Investigator), George A. Ostapowicz, M.D., Frank V. Schiødt, M.D., Julie Polson, M.D., University of Texas Southwestern, Dallas, TX; Anne M. Larson, M.D., Iris Liou, M.D., University of Washington, Seattle, WA; Timothy Davern, M.D., University of California, San Francisco, CA (current address: California Pacific Medical Center, San Francisco, CA), Oren Fix, M.D., University of California, San Francisco; Michael Schilsky, M.D., Mount Sinai School of Medicine, New York, NY (current address: Yale University, New Haven, CT); Timothy McCashland, M.D., University of Nebraska, Omaha, NE; J. Eileen Hay, M.B.B.S., Mayo Clinic, Rochester, MN; Natalie Murray, M.D., Baylor University Medical Center, Dallas, TX; A. Obaid S. Shaikh, M.D., University of Pittsburgh, Pittsburgh, PA; Andres Blei, M.D., Northwestern University, Chicago, IL (deceased), Daniel Ganger, M.D., Northwestern University, Chicago, IL; Atif Zaman, M.D., University of Oregon, Portland, OR; Steven H.B. Han, M.D., University of California, Los Angeles, CA; Robert Fontana, M.D., University of Michigan, Ann Arbor, MI; Brendan McGuire, M.D., University of Alabama, Birmingham, AL; Raymond T. Chung, M.D., Massachusetts General Hospital, Boston, MA; Alastair Smith, M.B., Ch.B., Duke University Medical Center, Durham, NC; Robert Brown, M.D., Cornell/Columbia University, New York, NY; Jeffrey Crippin, M.D., Washington University, St Louis, MO; Edwin Harrison, Mayo Clinic, Scottsdale, AZ; Adrian Reuben, M.B.B.S., Medical University of South Carolina, Charleston, SC; Santiago Munoz, M.D., Albert Einstein Medical Center, Philadelphia, PA; Rajender Reddy, M.D., University of Pennsylvania, Philadelphia, PA; R. Todd Stravitz, M.D., Virginia Commonwealth University, Richmond, VA; Lorenzo Rossaro, M.D., University of California Davis, Sacramento, CA; Raj Satyanarayana, M.D., Mayo Clinic, Jacksonville, FL; and Tarek Hassanein, M.D., University of California, San Diego, CA. The University of Texas Southwestern Administrative Group included Grace Samuel, Ezmina Lalani, Carla Pezzia, and Corron Sanders, Ph.D., Nahid Attar, the Statistics and Data Management Group included Joan S. Reisch, Ph.D., Linda S. Hynan, Ph.D., Janet P. Smith, Joe W. Webster and Mechelle Murray, and the Medical University of South Carolina Data Coordination Unit included Valerie Durkalski, Ph.D., Wenle Zhao, Ph.D., Catherine Dillon, and Tomoko Goddard.
Grant support: We gratefully acknowledge the support provided by the members of The Acute Liver Failure Study Group. This study was funded by the National Institute of Diabetes, Digestive and Kidney Diseases (DK U-01-58369). Additional funding provided by the Tips Fund of Northwestern Medical Foundation and the Jeanne Roberts and Rollin and Mary Ella King Funds of the Southwestern Medical Foundation. Additionally, Dr. Rangnekar is supported by the T32 DK62708-01, NIDDK Training Grant in Gastrointestinal Epidemiology, and a Clinical and Translational Science Award from the Michigan Institute for Clinical and Health Research
Glossary
- ALF
Acute Liver Failure
- ALFSG
Acute liver failure study group
- ALT
Alanine aminotransferase
- APAP
Acetaminophen
- DILI
Drug induced liver injury
- HBV
Hepatitis B virus
- ICP
Intracranial pressure
- IDU
Intravenous drug use
- INR
International normalized ratio
- LT
liver transplant
- MELD
Model for end-stage Liver disease
- SS
Spontaneous survivors
Footnotes
Author contributions:
Amol Rangnekar: study concept, analysis and interpretation, manuscript drafting and finalization.
Caitlyn Ellerbe: manuscript drafting and finalization, statistical analyses and interpretation.
Valerie Durkalski: study concept, manuscript drafting and finalization, statistical analyses and interpretation.
Robert J. Fontana: study concept, design, data acquisition, analysis and interpretation, manuscript drafting and finalization, overall supervision.
Brendan McGuire: Contribution of data, manuscript review and finalization
William M. Lee: manuscript drafting and finalization
Linda Hynan: final manuscript review including statistical analyses and interpretation
Rajender Reddy, Todd Stravitz, Timothy Davern, Adrian Reuben, Iris Liou, Oren Fix, Dan Ganger, Ray Chung, Mike Schilsky, Steve Han: contribution of data, manuscript review and finalization.
Disclosures: Drs. Rangnekar, Sanders, Hynan, Reuben, Liou, Han, Schilsky, Reuben, Stravitz, Fix, Davern, and Durkalski and Ms Ellerbe have no financial conflicts of interest. Dr. Fontana has served as a consultant to Bristol-Meyers Squibb, Vertex Pharmaceuticals, Tibotec, Merck, GlaxoSmithkline and Medtronic in the past year.
Reddy Consultant: Vertex, Janssen, Merck, Genentech-Roche, BMS, Idenix, Gilead, Novartis Investigator: Vertex, Janssen, BMS, Abbott, Ikaria
McGuire Grant support (Bayer, Cumberland, Ikaria, Vertex), consultant (Hepahope)
Ganger Consultant (speaker) to Vertex, Gilead, and Otsuka.
Chung Grant support (Gilead, Mass Biologics, Merck), consultant (Enanta, Idenix)
Lee Consultant (Lilly, Novartis, Cumberland) research support (BMS, Gilead, Merck, Roche, Boeringher- Ingleheim, Andays, Vertex)
Contributor Information
Caitlyn Ellerbe, Email: ellerbcn@musc.edu.
Valerie E. Durkalski, Email: durkalsv@musc.edu.
Amol Rangnekar, Email: amrangnekar@gmail.com.
K. Rajender Reddy, Email: rajender.reddy@uphs.upenn.edu.
Todd Stravitz, Email: rstravit@hsc.vcu.edu.
Brendan McGuire, Email: bmcguire@uab.edu.
Timothy Davern, Email: DavernT@sutterhealth.org.
Adrian Reuben, Email: reubena@musc.edu.
Iris Liou, Email: IrisL@medicine.washington.edu.
Oren Fix, Email: oren.fix@ucsf.edu.
Daniel R Ganger, Email: dganger@nmff.org.
Raymond T. Chung, Email: rtchung@partners.org.
Mike Schilsky, Email: Michael.schilsky@yale.edu.
Steven Han, Email: steven.han@ucla.edu.
Linda S. Hynan, Email: linda.hynan@utsouthwestern.edu.
Corron Sanders, Email: csand7@mednet.swmed.edu.
William M. Lee, Email: lee03@utsw.sw.med.edu.
References
- 1.Stravitz RT, Kramer AH, Davern T, et al. Intensive care of patients with acute liver failure: Recommendations of the US Acute Liver Failure Study Group. Crit Care Med. 2007;35:2498–2508. doi: 10.1097/01.CCM.0000287592.94554.5F. [DOI] [PubMed] [Google Scholar]
- 2.Ostapowicz G, Fontana RJ, Schiodt FV, et al. Results of a prospective study of acute liver failure at 17 tertiary care centers in the United States. Ann Intern Med. 2002;137:947–954. doi: 10.7326/0003-4819-137-12-200212170-00007. [DOI] [PubMed] [Google Scholar]
- 3.O’Grady JG, Alexander GJ, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology. 1989;97:439. doi: 10.1016/0016-5085(89)90081-4. [DOI] [PubMed] [Google Scholar]
- 4.Bernal W, Donaldson N, Wyncoll D, Wendon J. Blood lactate as an early predictor of outcome in paracetamol-induced acute liver failure: A cohort study. Lancet. 2002;359:558–563. doi: 10.1016/S0140-6736(02)07743-7. [DOI] [PubMed] [Google Scholar]
- 5.Freeman RB, Steffick DE, Guidinger MK, Farmer DG, Berg CL, Merion RM. Liver and intestine transplantation in the United States, 1997–2006. Am J Transplant. 2008;8:958–976. doi: 10.1111/j.1600-6143.2008.02174.x. [DOI] [PubMed] [Google Scholar]
- 6.Mohamed R, Hubscher SG, Mirza DF, Gunson BK, Mutimer DJ. Postransplantation chronic hepatitis in fulminant hepatic failure. Hepatology. 1997;25:1003–1007. doi: 10.1002/hep.510250435. [DOI] [PubMed] [Google Scholar]
- 7.Devlin J, Williams R. Transplantation for fulminant hepatic failure. Transplantation. 1996;62:151–155. doi: 10.1097/00007890-199611150-00012. [DOI] [PubMed] [Google Scholar]
- 8.Jackson EW, Zacks S, Zinn S, et al. Delayed neuropsychological dysfunction after liver transplantation for acute liver failure: A matched case-controlled study. Liver Transpl. 2002;10:932–936. doi: 10.1053/jlts.2002.35550. [DOI] [PubMed] [Google Scholar]
- 9.Lewis MB, Howdie PD. Cognitive dysfunction and HRQOL in long-term liver transplant survivors. Liver Transpl. 2003;9:1145–1148. doi: 10.1053/jlts.2003.50239. [DOI] [PubMed] [Google Scholar]
- 10.Taylor R, Tujios S, Jinjuvadia K, et al. Short and Long-term Outcomes in patients with acute liver failure due to ischemic hepatitis. Dig Dis Sciences. 2012;57:777–784. doi: 10.1007/s10620-011-1918-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Stravitz RT, Lefkowitch JH, Fontana RJ, et al. Autoimmune acute liver failure: Proposed clinical and histological criteria. Hepatology. 2010;53:517–526. doi: 10.1002/hep.24080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hanley JA, McNeil B. The meaning and use of the area under an ROC curve. Radiology. 1982;143:29–36. doi: 10.1148/radiology.143.1.7063747. [DOI] [PubMed] [Google Scholar]
- 13.Myers RP, Shaheen AM, Li B, Dean S, Quan H. Impact of liver disease, alcohol abuse, and unintentional ingestions on the outcomes of acetaminophen overdose. Clin Gastro Hep. 2008;6:918–926. doi: 10.1016/j.cgh.2008.02.053. [DOI] [PubMed] [Google Scholar]
- 14.Germani G, Theocharidou E, Adam R, et al. Liver Transplantation for acute liver failure in Europe: Outcomes over 20 years from the ELTR database. J Hepatology. 2012;57:288–296. doi: 10.1016/j.jhep.2012.03.017. [DOI] [PubMed] [Google Scholar]
- 15.Bernal W, Cross TJ, Auzinger G, et al. Outcome after wait-listing for emergency liver transplantation in acute liver failure: A single center experience. J Hepatol. 2009;50:306–313. doi: 10.1016/j.jhep.2008.09.012. [DOI] [PubMed] [Google Scholar]
- 16.Barshes NR, Lee TC, Balkrishnan R, Karpen SJ, Carter BA, Goss JA. Risk stratification of adult patients undergoing liver transplantation for fulminant hepatic failure. Transplantation. 2006;81:195–201. doi: 10.1097/01.tp.0000188149.90975.63. [DOI] [PubMed] [Google Scholar]
- 17.Yamashiki N, Sugawara Y, Tamura S, et al. Outcomes after living donor liver transplantation for acute liver failure in Japan: Results of a Nationwide Survey. Liver Transpl. 2012;18:1069–1077. doi: 10.1002/lt.23469. [DOI] [PubMed] [Google Scholar]
- 18.Schmidt LE, Dalhoff K. Serum phosphate is an early predictor of outcome in severe acetaminophen-induced hepatotoxicity. Hepatology. 2002;36:659–663. doi: 10.1053/jhep.2002.35069. [DOI] [PubMed] [Google Scholar]
- 19.Macquillan GC, Seyam MS, Nightingale P, Neuberger JM, Murphy N. Blood lactate but not serum phosphate levels can predict patient outcome in fulminant hepatic failure. Liver Transpl. 2005;11:1073–1079. doi: 10.1002/lt.20427. [DOI] [PubMed] [Google Scholar]
- 20.Chan G, Taqui A, Marotta P, Levstik M, McAlister V, Wall W, Quan D. Long-term outcomes of emergency liver transplantation for acute liver failure. Liver Transpl. 2009;15:1696–1702. doi: 10.1002/lt.21931. [DOI] [PubMed] [Google Scholar]
- 21.Tan WF, Steadman RH, Farmer DG, Hong JC, Busuttil RW, Apinyachon W, Xia VW. Pretransplant neurological presentation and severe posttransplant brain injury in patients with acute liver failure. Transplantation. 2012;94:768–774. doi: 10.1097/TP.0b013e3182620596. [DOI] [PubMed] [Google Scholar]
- 22.Forde KA, Reddy KR, Troxel AB, Sanders CM, Lee WM. Racial and ethnic differences in presentation, etiology, and outcomes of acute liver failure in the US. Clin Gastro Hep. 2009;7:1121–1126. doi: 10.1016/j.cgh.2009.05.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bjornsson E, Olsson R. Outcome and prognostic markers in severe Drug-induced liver Disease. Hepatology. 2005;42:481–489. doi: 10.1002/hep.20800. [DOI] [PubMed] [Google Scholar]; Denie SC, Douarin L, Francoz C, et al. Levels and initial course of serum alanine aminotransferase can predict outcome of patients with Budd-Chiari syndrome. Clin Gastro Hep. 2009;7:1236–1240. doi: 10.1016/j.cgh.2009.06.011. [DOI] [PubMed] [Google Scholar]
- 24.Gimson AE, O’Grady J, Ede RJ, Portman B, Williams R. Late-onset hepatic failure: Clinical, serological and histological features. Hepatology. 1986;6:288. doi: 10.1002/hep.1840060222. [DOI] [PubMed] [Google Scholar]
- 25.Dhiman RK, Makharia GK, Jain S, Chawla Y. Ascites and spontaneous bacterial peritonitis in fulminant hepatic failure. Am J Gastroenterol. 2000;95:233. doi: 10.1111/j.1572-0241.2000.01691.x. [DOI] [PubMed] [Google Scholar]
- 26.Vaquero J, Polson J, Chung C, et al. Infection and the progression of hepatic encephalopathy in acute liver failure. Gastroenterology. 2003;125:755–764. doi: 10.1016/s0016-5085(03)01051-5. [DOI] [PubMed] [Google Scholar]
- 27.McDiarmid SV, Goodrich NP, Harper AM, Merion RM. Liver Transplantation for Status 1: The consequences of Good Intentions. Liver Transpl. 2007;13:699–707. doi: 10.1002/lt.21125. [DOI] [PubMed] [Google Scholar]
- 28.Larson AM, Polson J, Fontana RJ, et al. Acetaminophen-induced acute liver failure: results of a United States multicenter, prospective study. Hepatology. 2005;42:1364–72. doi: 10.1002/hep.20948. [DOI] [PubMed] [Google Scholar]
- 29.Aberg F, Hockerstedt K, Roine RP, Sintonen H, Isoniemi H. Influence of liver-disease etiology on long-term quality of life and employment after liver transplantation. Clin Transplant. 2012;26:729–735. doi: 10.1111/j.1399-0012.2012.01597.x. [DOI] [PubMed] [Google Scholar]
- 30.Guimaro MS, Lacerda SS, Aguilar MR, Karam CH, Kernkraut AM, Ferraz-Neto BH. Post-traumatic stress disorders, mood disorders, and quality of life in transplant recipients with acute liver failure. Transplant Proc. 2011;43:187–188. doi: 10.1016/j.transproceed.2010.11.015. [DOI] [PubMed] [Google Scholar]
- 31.Rangnekar AS, Ellerbe C, Durkalski V, McGuire B, Lee WM, Fontana RJ. Quality of life is significantly impaired in long-term survivors of acute liver failure and particularly in acetaminophen overdose patients. Liver Transplantation. 2013;19:991–1000. doi: 10.1002/lt.23688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Helweg-Larsen K. The Danish Register of Causes of Death. Scan J Pub Health. 2011;39 (Suppl 7):26–29. doi: 10.1177/1403494811399958. [DOI] [PubMed] [Google Scholar]
- 33.Potterat JJ, Brewer DD, Muth SQ, et al. Mortality in a long-term open cohort of Prostitute women. Am J Epidemiology. 2004;159:778–785. doi: 10.1093/aje/kwh110. [DOI] [PubMed] [Google Scholar]
- 34.Reidel D, Mwangi E, Fantry L, et al. High cancer-related mortality in an urban, predominantly African-American, HIV-infected population. AIDS. 2013;27:1109–1107. doi: 10.1097/QAD.0b013e32835dc068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Asnaes S. Uncertainity of determining mode and cause of death without autopsy: An autopsy study of medically unattended non-medicolegal deaths. For Sci Int. 1980;15:191–196. doi: 10.1016/0379-0738(80)90133-4. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.