shows the overall outcomes for all patients included in the study. The spontaneous (ie, transplantation-free) survival rate, the liver transplantation rate, and the overall survival rate were 46.0%, 25.2%, and 68.4%, respectively.
Overall outcome for all acute liver failure patients.
shows the age distribution of the patients. The median age was 38 years (range, 15–81 years). One thousand sixteen patients (90.2%) were younger than 60 years and composed group 1. The 110 patients ≥ 60 years composed group 2. Four hundred ninety-nine patients in group 1 (49.1%) had acetaminophen-induced ALF; this rate of acetaminophen-induced ALF was significantly higher than that in group 2 (22.7%, P < 0.001). Older patients had lower aminotransferases on admission and apparently lower spontaneous survival rates in comparison with younger patients ().
Age distribution for all patients. Abbreviation: ACM, acetaminophen.
Overall Comparison of Younger and Older Patients
For acetaminophen patients, groups 1 and 2 are compared in . There was a tendency toward higher encephalopathy grades in group 2. Fewer patients in group 2 were listed for transplantation. The gender distribution, symptom duration, liver transplantation rate, biochemical markers (except aminotransferates, and spontaneous survival rates did not differ between the 2 age groups. (upper panel) further shows the spontaneous survival rate, and (upper panel) shows the liver transplantation rate for all age groups with an acetaminophen etiology.
Comparison of Younger and Older Patients with an Acetaminophen Etiology (n = 524)
Spontaneous survival rates for patients with an ACM etiology (upper panel) or a non-ACM (lower panel) etiology as a function of age. With linear regression analysis, no trends were found. Abbreviation: ACM, acetaminophen.
Transplantation rates for patients with an ACM etiology (upper panel) or a non-ACM etiology as a function of age. With linear regression analysis, a decrease in transplantation was observed with increasing age. Abbreviation: ACM, acetaminophen.
compares groups 1 and 2 for non-acetaminophen patients. The non-acetaminophen group (n = 602) consisted of those with drug-induced liver injury (n = 127; 21.1%), hepatitis B (n = 82; 13.6%), autoimmune hepatitis (n = 60; 10.0%), shock liver (n = 47; 7.8%), hepatitis A (n = 31; 5.1%), Wilson’s disease (n = 18; 3.0%), Budd-Chiari syndrome (n = 11; 1.8%), an indeterminate etiology (n = 158; 26.2%), and various other etiologies (n = 68; 11.3%). Pregnancy-related ALF, Budd-Chiari syndrome, and Wilson’s disease were not observed in the older age group, and autoimmune hepatitis was less frequent in the younger age group (8.7% versus 17.6%, respectively, P = 0.026). Fewer patients were listed for transplantation in group 2, and more patients had contraindications for transplantation in this group, including 5 patients for whom old age was listed as a cause for contraindication. These 5 patients were older than 70 years of age. The gender distribution, hepatic coma grade, symptom duration, biochemical markers, and spontaneous survival did not differ between the groups, whereas there was a tendency toward a lower transplantation rate in group 2. Spontaneous survival rates did not differ for age groups, whereas transplantation rates decreased with older age according to linear regression analysis ( and , lower panels).
Comparison of Younger and Older Patients with a Non-Acetaminophen Etiology (n = 602)
Other contraindications for liver transplantation, besides age, included comorbidities, the medical condition of the patient (eg, uncontrollable infection, multi-organ failure, or severe intracranial hypertension), psychiatric disorders, chronic alcohol use, repeated overdoses, and malignancy and did not differ between acetaminophen and non-acetaminophen patients.
Survival after liver transplantation was not age-dependent ().
Figure 5 Survival after transplantation for 48 patients with ACM-induced acute liver failure and for 236 patients with a non-ACM etiology according to age. No significant differences were found. Abbreviations: ACM, acetaminophen; NA, not available; Tx, transplantation. (more ...)
Age was not a significant predictor of spontaneous survival in multiple logistic regression analyses of outcome for the acetaminophen and non-acetaminophen subgroups with the following variables: age, admission or maximum international normalized ratio, symptom duration, admission bilirubin, creatinine, phosphate, magnesium, acidosis, and hepatic encephalopathy grade (I/II versus III/IV). When transplanted patients were excluded from the analysis, old age was an independent risk factor for nonsurvival in the non-acetaminophen subgroup (P = 0.018), whereas no impact of age was found in the acetaminophen group.
Age > 40 years is a significant risk factor for prognosis in the King’s College Hospital criteria10
; therefore, we analyzed outcome with this cutoff level. There were more acetaminophen patients in the group younger than 40 years (n = 322; 53.4%) versus the group of patients 40 years old or older (n = 202; 38.6%, P
< 0.001). In the acetaminophen group, the spontaneous survival rate was higher in the younger group than in the older group (68.0% versus 58.9%, respectively, P
= 0.043), whereas overall survival and transplantation rates did not differ between the 2 groups. In the non-acetaminophen group, the overall survival was higher in the younger group than in the older group (70.1% versus 60.7%, respectively, P
= 0.026). Also, the transplantation rate was higher in the younger group than in the older group (44.1% versus 34.9%, respectively, P
= 0.026), whereas the spontaneous survival rate did not differ in the 2 groups.
We previously reported rises in alpha-fetoprotein levels as a sign of hepatic regeneration,22
and we analyzed these data again with respect to age; among nonsurvivors (or transplanted) older than 60 years, 6 of 13 (46%) patients had rising alpha-fetoprotein levels between hospital days 1 and 3 versus only 7 of 51 (13%) patients younger than 60 years (P
= 0.018). Among spontaneous survivors, the proportion of patients with alpha-fetoprotein rise did not differ between older and younger patients (57% and 73%, respectively; not significant).