Cohort characterization at time of biopsy
A total of 43 patients with ALF receiving TJLB were included into the study (Table 1). The median (interquartile range (IQR)) age was 55 (37-64) years and the majority of 30 patients (69.8%) were female. At time of TJLB, all patients had manifest hepatic encephalopathy (HE), with grade I HE in 16 (37.2%), grade II HE in five (11.6%) and grade III HE in 22 (51.2%) patients, respectively. At this time, 17 (39.5%) patients were already admitted to the ICU, one patient (2.3%) was invasively ventilated, five patients (11.6%) were receiving renal replacement therapy (RRT) and ten patients (23.2%) therapeutic plasma exchange (TPE). The median (IQR) MELD-Score was 30 (27-35) points. Autoimmune hepatitis (AIH) antibodies were detectable before biopsy obtainment in the majority of patients, e.g. antinuclear antibodies (ANA) in 26 (60.4%) and smooth-muscle (SMA) in 28 (65.1%) patients, respectively. Median ANA and SMA-titers were 1:160 and 1:80, respectively.
Within 28 days following TJLB, 14 (32.6%) patients died. 15 (34.9%) patients were HU listed for LTX and 14 (32.6%) received a LTX.
Table 1: Demographic and clinical parameters at time of transjugular liver biopsy
|
N (%)/Median (IQR)
|
Age, years
|
55 (37-64)
|
BMI, kg/m2
|
24.4 (21.9-28.6)
|
Sex
|
|
female
|
30 (69.8)
|
male
|
13 (30.2)
|
HE
|
|
Grade I
|
16 (37.2)
|
Grade II
|
5 (11.6)
|
Grade III
|
22 (51.2)
|
Grade IV
|
0
|
ICU Admission
|
17 (39.5)
|
Invasive Ventilation
|
1 (2.3)
|
RRT
|
5 (11.6)
|
Vasopressor therapy
|
3 (7)
|
Therapeutic Plasma Exchange
|
10 (23.2)
|
INR
|
2.2 (1.9-3.4)
|
AST, U/l
|
1227 (452-1682)
|
ALT, U/l
|
1301 (408-2601)
|
Bilirubin, µmol/l
|
342 (186-446)
|
Creatinine, µmol/l
|
76 (59-161)
|
MELD, points
|
30 (27-35)
|
Lactate, mmol/l
|
2.5 (1.7-3.4)
|
Ammonia, µmol/l
|
60 (46-73)
|
Factor V, %
|
38 (19.5-48.2)
|
Autoimmune-Hepatitis Antibodies
|
|
ANA positive
|
26 (60.4)
|
ANA Titer
|
1:160 (1:80-1:160)
|
SMA
|
28 (65.1)
|
SMA Titer
|
1:80 (1:80-1:160)
|
LKM
|
6 (14)
|
LKM Titer
|
1:160
|
ABBREVIATIONS: ANA – Antinuclear antibodies, BMI – Body mass index, HU – High urgency, ICU – Intensive Care Unit, LKM – Liver-kidney microsomal antibodies, LTX – Liver transplantation, RRT – Renal replacement therapy, SMA – Smooth-muscle antibodies
Procedural and safety characteristics of transjugular liver biopsy
TJLB procedure related complications occurred in none of the patients. The technical success rate was 100%. The median (IQR) size of the TJLB specimen was 1.3 (0.3-3) cm. The biopsy was usable for further histopathological examination in all but two patients. In one patient biopsy was not representative, in the other too much intra-hepatic necrosis was present for further histopathological examination. The median (IQR) radiation dose applied was 1477 (551-3828) cGy/cm².
Impact of histopathological results obtained by transjugular liver biopsy for diagnosis of aetiology and further therapeutic approach
The most likely assumed aetiology of ALF and the therapeutic approach before- and after obtaining histopathological results by TJLB were compared in all patients by re-examining chart records (Table 2). 23 (53.4%) patients had suspected drug-induced liver failure with no change in diagnosis or therapeutic approach following biopsy. Five (11.6%) cases were initially categorized as cryptogenic, of which a definitive diagnosis was found in three of these patients and therapeutic measures were changed in only one (patient with histopathological diagnosis of AIH). Viral hepatitis was confirmed as the presumed aetiology in four (9.3%) patients with no change in the consecutive therapeutic measures. High suspicion for AIH as underlying aetiology was present in five (11.6%) patients and this diagnosis was confirmed by TJLB in four of these. Immunosuppression was started before TJLB in three of these patients and was continued as long-term immunosuppression in both following definitive diagnosis of AIH. In one presumed AIH aetiology TJLB could not confirm AIH but suggested a more likely drug induced cause and no immunosuppressive treatment was started. In another presumed AIH aetiology, histology was generally compatible with AIH, but hepatitis resolved without immunosuppressive therapy, which might have been in line with the new term of drug-induced autoimmune like hepatitis (16). In two patients with sickle cell crisis associated ALF vaso-occlusive aetiology was presumed and this could be confirmed in both by histology, indicating repetitive exchange transfusions. Hepatic manifestation of graft versus host disease (GvHD) needed histologic confirmation in one patient, then indicating high dose immunosuppression.
In summary, in most cases (n=39/43 cases) TJLB confirmed the initially already clinically presumed ALF aetiology and the therapeutic approach was unchanged by additional histological examination in the majority of patients (36/43 cases). However, in patients with a high suspicion for aetiologies potentially treatable by medical immunosuppression (e.g. AIH, GvHD), TJLB significantly influenced further treatment planning and/or adjustment in all of these patients.
Table 2: Significance of histopathological biopsy for clinical diagnosis and therapeutic approach
|
Most likely diagnosis before biopsy
|
Histopathological diagnosis
|
Diagnosis changed by biopsy
|
Therapeutic approach
influenced by biopsy
|
Therapeutic approach before biopsy
|
Therapeutic approach after therapy
|
autoimmune
|
drug-induced
|
Yes
|
Yes
|
supportive therapy
|
supportive therapy
|
autoimmune
|
autoimmune
|
No
|
Yes
|
immunosuppression
|
continuation of long-term immunosuppression
|
autoimmune
|
autoimmune
|
No
|
Yes
|
immunosuppression
|
continuation of long-term immunosuppression
|
autoimmune
|
autoimmune
|
No
|
Yes
|
immunosuppression
|
continuation of long-term immunosuppression
|
autoimmune
|
autoimmune
|
No
|
No
|
supportive therapy
|
no immunosuppression due to spontaneous recovery
|
cryptogenic
|
autoimmune
|
Yes
|
Yes
|
supportive therapy
|
immunosuppression
|
cryptogenic
|
drug-induced
|
Yes
|
No
|
supportive therapy
|
supportive therapy
|
cryptogenic
|
drug-induced
|
Yes
|
No
|
supportive therapy
|
supportive therapy
|
graft-versus-host disease
|
graft-versus-host disease
|
No
|
Yes
|
immunosuppression
|
continuation of long-term immunosuppression
|
vaso-occlusive (sickle cell crisis)
|
vaso-occlusive (sickle cell crisis)
|
No
|
Yes
|
exchange transfusion
|
continuation of exchange transfusion
|
vaso-occlusive (sickle cell crisis)
|
not representative
|
No
|
No
|
exchange transfusion
|
continuation of exchange transfusion
|
alcoholic
|
steatohepatitis
|
No
|
No
|
supportive therapy
|
supportive therapy
|
alcoholic
|
steatohepatitis
|
No
|
No
|
supportive therapy
|
supportive therapy
|
cryptogenic
|
unclear
|
No
|
No
|
supportive therapy
|
supportive therapy
|
cryptogenic
|
unclear
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
unclear
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
drug-induced
|
drug-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
HAV infection
|
virus infection
|
No
|
No
|
supportive therapy
|
supportive therapy
|
HBV infection
|
too much intrahepatic necrosis
|
No
|
No
|
antiviral therapy
|
antiviral therapy
|
HBV infection
|
virus infection
|
No
|
No
|
antiviral therapy
|
antiviral therapy
|
HEV infection
|
virus infection
|
No
|
No
|
antiviral therapy
|
antiviral therapy
|
pregnancy-induced
|
pregnancy-induced
|
No
|
No
|
supportive therapy
|
supportive therapy
|
ABBREVIATIONS: HAV – Hepatitis A virus, HBV – Hepatitis B Virus, HEV – Hepatitis E Virus
Predictors of liver-transplant free survival
The predictive value of various clinical and laboratory characteristics at the time of TJLB as well as histopathological findings obtained by TJLB, e.g. the Ishak Score categories and the degree of intrahepatic necrosis, on liver transplant-free survival were evaluated (Table 3). In the univariate analysis, the MELD Score (p=0.06), blood lactate concentration (p=0.02), pH (p=0.06), the presence of HE greater than I° (p =0.08), the BiLE Score (p=0.002) and the need for vasopressor therapy (p=<0,01) were associated with liver-transplant free survival. None of the Ishak subscore categories showed a predictive value on liver-transplant free survival. While the majority of 31 patients (72%) had significant intrahepatic necrosis according to the Ishak score, the exact histopathological degree of intrahepatic necrosis in the TJLB specimen varied substantially in individual patients (Median (IQR): 60 (10-90)). While the degree of intrahepatic necrosis showed significance in the univariate analysis (p=0.04), it did not demonstrate a significant predictive effect on liver transplant-free survival in the multivariate analysis (p=0.1). In fact, the BiLE Score was the only parameter at the time of TJLB performance, that had a significant effect on transplant free survival (Hazard Ratio (HR) 1.2 (95% Confidence Interval (CI) 1.06-1.3), p=0.002). Patients with acute intrahepatic necrosis of below and above 50% of the liver parenchyma had comparable liver-transplant free survival rates (Hazard Ratio (HR) 1.38 (95% Confidence Interval (CI) 0.5-3.9), p=0.527, Figure 2A). In contrast, a previously suggested cut-off value for the BilE Score (15) determined at the time of TJLB procedure, sufficiently predicted later liver-transplant free survival (Hazard Ratio (HR) 2.65 (95% Confidence Interval (CI) 1.02-6.88), p=0.045, Figure 2B).
Table 3: Predictors of liver transplant-free survival at time of transjugular liver biopsy
|
|
|
univariate
|
|
|
multivariate
|
|
|
HR
|
95% CI
|
p
|
HR
|
95% CI
|
p
|
MELD Score, points
|
1,08
|
1 – 1,17
|
0,06
|
1
|
0,89 – 1,12
|
0,92
|
Lactate, mmol/l
|
1,15
|
1,02 – 1,3
|
0,02
|
0,56
|
0,29 – 1,06
|
0,08
|
pH
|
0,01
|
0,00 – 1,4
|
0,07
|
0
|
0,00 – 16,78
|
0,2
|
Ammonia, µmol/l
|
0,99
|
0,96 -1
|
0,5
|
|
|
|
Factor V, %
|
1
|
0,97 – 1,03
|
1
|
|
|
|
LDH, U/l
|
1
|
0,99 - 1
|
0,17
|
|
|
|
HE greater than I°
|
2,31
|
0,92 – 5,9
|
0,08
|
1,08
|
0,30 – 3,81
|
0,9
|
BiLE Score, points
|
1,2
|
1,06 – 1,3
|
0,002
|
1,5
|
0,99 – 2,33
|
0,053
|
ICU admission
|
1,7
|
0,8 – 3,8
|
0,2
|
|
|
|
Invasive Ventilation
|
0,05
|
0 – 1110
|
0,6
|
|
|
|
RRT
|
0,65
|
0,16 – 2,75
|
0,5
|
|
|
|
Vasopressor therapy
|
6
|
1,6 – 20
|
< 0,01
|
14
|
0,27 – 713,41
|
0,19
|
HVPG, mmHg
|
1
|
0,9 – 1,3
|
0,3
|
|
|
|
Ishak-Score
|
|
|
|
|
|
|
Ishak A
|
1,14
|
0,78 – 1,7
|
0,5
|
|
|
|
Ishak B
|
1,3
|
0,9 – 1,8
|
0,17
|
|
|
|
Ishak C
|
1
|
0,6– 1,7
|
0,9
|
|
|
|
Ishak D
|
1,5
|
0,9 – 2,5
|
0,14
|
|
|
|
Ishak F
|
0,8
|
0,5 – 1,1
|
0,2
|
|
|
|
Intrahepatic necrosis, %
|
1,03
|
1,03 - 1
|
0,04
|
1,04
|
0,99– 1,08
|
0,1
|
ABBREVIATIONS: CI – Confidence interval, HE – Hepatic encephalopathy, HR – Hazard ratio, HVPG – Hepatic venous pressure gradient, ICU –Intensive care unit, LDH – Lactate dehydrogenase, MELD – Model for end stage liver disease, RRT – Renal replacement therapy
Impact of histopathological degree of intrahepatic necrosis on further secondary clinical endpoints
It was also investigated whether the histological extent of intrahepatic necrosis had a significant impact on further secondary clinical endpoints (Table 4). However, degree of intrahepatic necrosis was neither associated with later need for intubation and invasive ventilation, RRT or HU listing for LTX. Only consecutive ICU admission was more likely with higher extent of intrahepatic necrosis (Odds ratio (OR) 1.04 (95% Confidence Interval (CI) 1-1.08), p=0.046). Another potential strength of transjugular liver biopsy (TJLB) may lie in its ability to exclude advanced fibrosis through histopathological analysis. This allows for definitive confirmation of acute liver failure (ALF) and exclusion of acute-on-chronic liver failure (ACLF).
Table 4: Impact of histopathological degree of intrahepatic necrosis on secondary clinical endpoints
|
OR
|
95% CI
|
p
|
|
|
|
|
ICU admission
|
1.04
|
1 – 1.08
|
0.046
|
Intubation
|
1
|
0,99 – 1.06
|
0.2
|
RRT
|
1
|
1 – 1.05
|
0.6
|
HU listing for LTX
|
1.04
|
1 – 1.09
|
0.07
|
ABBREVIATIONS: CI – Confidence interval, HU – High-urgency, ICU – Intensive care unit, LTX – Liver transplantation, OR – Odds ratio, RRT – Renal replacement therapy