About the Chronic Liver Failure — Sequential Organ Failure Assessment (CLIF-SOFA)
What CLIF-SOFA is actually for is identifying organ failures, not adding points. It adapts the intensive care SOFA score to cirrhosis across six systems — liver, kidney, coagulation, circulation, brain and lungs — and the total matters far less than which of them have crossed their failure threshold. Two or more failures give ACLF grade 2 or 3; a single failure only qualifies as grade 1 if it is kidney failure, or is accompanied by creatinine 1.5–1.9 mg/dL or hepatic encephalopathy grade I–II. In the CANONIC study of 1,343 patients, 28-day mortality was 22.1% at grade 1, 32.0% at grade 2 and 76.7% at grade 3, against 1.9% with no organ failure at all.
Formula
CLIF-SOFA = liver + kidney + coagulation + circulation + brain + lungs (each 0–4)
Organ failure = liver ≥ 3, kidney ≥ 2, coagulation ≥ 3, circulation ≥ 3, brain ≥ 3, lungs ≥ 3- liver
- Bilirubin in mg/dL: 0 below 1.2, 1 below 2, 2 below 6, 3 below 12, 4 at 12 or above. Failure at 3.
- kidney
- Creatinine in mg/dL: 0 below 1.2, 1 below 2, 2 below 3.5, 3 below 5, 4 at 5 or above. Failure at 2 — the only system whose failure threshold is a subscore of 2.
- coagulation
- INR: 0 below 1.1, 1 below 1.25, 2 below 1.5, 3 below 2.5, 4 at 2.5 or above. Failure at 3.
- circulation
- 3 if on vasopressors; otherwise 0 at a mean arterial pressure of 70 mmHg or more, 1 below 70. Failure at 3, which in practice means vasopressor dependence.
- brain
- West Haven grade 0 to 4 used directly as the subscore. Failure at 3 — West Haven grade III, somnolence to semi-stupor.
- lungs
- PaO₂/FiO₂: 0 above 400, 1 above 300, 2 above 200, 3 above 100, 4 at 100 or below. Or SpO₂/FiO₂: 0 above 512, 1 above 357, 2 above 214, 3 above 89, 4 at 89 or below. Mechanical ventilation forces at least 3. Failure at 3.
- The failure thresholds are not uniform. Five systems fail at a subscore of 3 and the kidney fails at 2, which is easy to miss and changes the grade.
- The total is not the output that matters. A CLIF-SOFA of 10 spread across six systems with none in failure is a different patient from a CLIF-SOFA of 10 concentrated in three failing systems.
- Vasopressor use overrides mean arterial pressure. A patient maintained at 75 mmHg on noradrenaline scores 3 and is in circulatory failure.
- Mechanical ventilation forces the respiratory subscore to at least 3, so an intubated patient is in respiratory failure whatever their oxygenation ratio shows.
- The SpO₂/FiO₂ thresholds are not the PaO₂/FiO₂ thresholds. They are separate cut-offs derived from the relationship between the two ratios, and using one set with the other measurement misclassifies patients.
- Grade 1 is not simply "one organ failure". It is single kidney failure, OR a single liver, coagulation, circulatory or respiratory failure accompanied by creatinine 1.5–1.9 mg/dL and/or hepatic encephalopathy grade I–II, OR single cerebral failure with creatinine 1.5–1.9 mg/dL. A single non-kidney organ failure with normal renal and cerebral function is organ failure without ACLF.
Interpreting the result
No organ failure in a patient admitted with decompensated cirrhosis carries a 28-day mortality of about 1.9%, and that patient does not have acute-on-chronic liver failure — treat the decompensation and watch, because ACLF can declare itself over the following days. Grade 1 carries roughly 22% 28-day mortality, grade 2 about 32%, and grade 3 about 77%. The single most useful thing to know about these grades is that they are dynamic: the CLIF-C ACLF score computed at 48 hours and again at three to seven days predicts 28-day mortality significantly better than the value at diagnosis, so a grade recorded on admission is a starting point rather than a verdict. Roughly two in five patients have no precipitant identified, which is worth knowing so that failure to find one does not stop the search prematurely — infection, alcohol, variceal bleeding and drug injury account for most of the rest and all are actionable. Assess transplant candidacy early, because the practical window narrows as organs fail and a grade 3 patient who was never assessed at grade 1 has usually lost the opportunity rather than been denied it.
| Score | Band | What it means | Action |
|---|---|---|---|
| No organ failure | Decompensated cirrhosis without ACLF | 28-day mortality approximately 1.9% in the CANONIC cohort | Treat the decompensation and its precipitant; reassess daily, since 112 of 1,343 CANONIC patients developed ACLF after enrolment |
| 1 organ failure meeting the qualifying pattern | ACLF grade 1 | 28-day mortality 22.1% | Identify and treat the precipitant; intensive monitoring; begin transplant assessment now rather than on deterioration |
| 2 organ failures | ACLF grade 2 | 28-day mortality 32.0% | Critical care input; reassess the grade at 48 hours and at 3–7 days, since trajectory outperforms the admission value |
| 3 or more organ failures | ACLF grade 3 | 28-day mortality 76.7% | Intensive care and urgent transplant evaluation; where transplantation is not an option, involve palliative care alongside active treatment |
Scroll the table sideways for every column.
What the CLIF-SOFA needs (6 inputs)
- Total bilirubin — liver
- Scored 0 below 1.2 mg/dL, 1 to 2, 2 to 6, 3 to 12, and 4 above 12. Liver failure is defined at a subscore of 3, meaning bilirubin of 12 mg/dL or above.
- Creatinine — kidney
- Scored 0 below 1.2 mg/dL, 1 to 2, 2 to 3.5, 3 to 5, and 4 above 5. Kidney failure is defined at a subscore of 2 — a creatinine of 2.0 mg/dL — which is a lower bar than any other system and is deliberate, because renal dysfunction in cirrhosis carries disproportionate prognostic weight.
- INR — coagulation
- Scored 0 below 1.1, rising through 1.25 and 1.5 to 4 at 2.5 and above. Coagulation failure is a subscore of 3, meaning an INR of 2.5 or above.
- Mean arterial pressure and vasopressor use — circulation
- 0 at a mean arterial pressure of 70 mmHg or above, 1 below 70. Vasopressor use overrides the pressure entirely and scores 3, which is the circulatory failure threshold — so a patient with a normal pressure on noradrenaline is in circulatory failure.
- Hepatic encephalopathy grade — brain
- The West Haven grade entered directly as the subscore, 0 to 4. Cerebral failure is grade 3 or above. An error in West Haven grading propagates straight into the ACLF grade.
- PaO₂/FiO₂ or SpO₂/FiO₂ ratio, and mechanical ventilation — lungs
- Either ratio can be used; the arterial one is preferred where a blood gas is available. Mechanical ventilation sets the respiratory subscore to at least 3 regardless of the ratio, which is the respiratory failure threshold.
Units. Bilirubin and creatinine can be entered in either conventional or SI units and are converted internally. The published thresholds are in mg/dL: bilirubin failure at 12 mg/dL (205 µmol/L) and kidney failure at a creatinine of 2.0 mg/dL (177 µmol/L). The respiratory subscore has two separate threshold sets — PaO₂/FiO₂ at 400/300/200/100 and SpO₂/FiO₂ at 512/357/214/89 — and they are not interchangeable; applying the arterial cut-offs to a saturation-derived ratio will misclassify the subscore in either direction.
What it returns
- CLIF-SOFA total
- The sum of the six subscores. Reported for completeness, but the total is not what the ACLF definition uses and two patients with identical totals can differ by two grades.
- Number of organ failures
- The count of systems at or above their failure threshold, and which ones. Two or more settles the grade on its own; a single failure has to be tested against the grade 1 qualifying patterns.
- ACLF grade and 28-day mortality
- Grade 1, 2 or 3 with the CANONIC 28-day mortality attached: 22.1%, 32.0% and 76.7% respectively.
How it is calculated
The score is an adaptation of the intensive care SOFA score, retuned so that its thresholds reflect what the same laboratory values mean in a cirrhotic patient rather than a general critical care population. But the important methodological step in CANONIC was not the score — it was how the syndrome was defined from it. The investigators enrolled 1,343 patients hospitalised for acute decompensation of cirrhosis across 29 liver units in eight European countries, recorded organ failures by CLIF-SOFA, and then asked which patterns of failure identified a group with a 28-day mortality above 15%. Acute-on-chronic liver failure was defined as those patterns. The grading follows the same logic: mortality rises steeply with the number of failing organs, and the three grades mark that gradient. This is why the framework is empirical rather than consensus-derived, and why the renal threshold sits lower than the others — the data, not clinical opinion, put it there.
Facts & figures
| System | Subscore defining failure | What that means clinically |
|---|---|---|
| Liver | ≥ 3 | Bilirubin ≥ 12 mg/dL |
| Kidney | ≥ 2 | Creatinine ≥ 2.0 mg/dL — the lowest bar of the six, by design |
| Coagulation | ≥ 3 | INR ≥ 2.5 |
| Circulation | ≥ 3 | Vasopressor requirement, irrespective of the achieved blood pressure |
| Brain | ≥ 3 | West Haven grade III or IV |
| Lungs | ≥ 3 | PaO₂/FiO₂ ≤ 200, SpO₂/FiO₂ ≤ 214, or mechanical ventilation |
Scroll the table sideways for every column.
Reading all six as failing at 3 is the commonest error and systematically under-diagnoses ACLF, because renal dysfunction is both the most frequent organ failure in this population and the one with the lowest threshold.
| Group | n | 28-day mortality |
|---|---|---|
| ACLF at enrolment | 303 | 33.9% |
| Developed ACLF during follow-up | 112 | 29.7% |
| Never developed ACLF | 928 | 1.9% |
Scroll the table sideways for every column.
The 112 patients who developed ACLF after enrolment are the reason a single admission score is insufficient. Their mortality was close to that of patients who presented with the syndrome, and none of them would have been identified by an admission assessment alone.
Evidence
Derivation — CANONIC study, EASL-CLIF Consortium
2013 · n = 1,343Prospective observational study of patients hospitalised for acute decompensation of cirrhosis at 29 liver units in eight European countries between February and September 2011. Diagnostic criteria for ACLF were established by identifying which patterns of organ failure, defined by CLIF-SOFA, carried a 28-day mortality above 15%.
303 patients had ACLF at enrolment (28-day mortality 33.9%), 112 developed it during follow-up (29.7%), and 928 never did (1.9%). By grade, 28-day mortality was 22.1% at grade 1, 32.0% at grade 2 and 76.7% at grade 3.
Simplification and prognostic score — CLIF-C OF and CLIF-C ACLF
2014 · n = 275Development and validation within the CANONIC dataset of a simplified organ function score (CLIF-C OF) for diagnosing ACLF, and a prognostic score (CLIF-C ACLF) built on 275 patients with ACLF.
CLIF-C ACLF showed significantly higher predictive accuracy than MELD, MELD-Na and Child-Pugh, reducing prediction error rates by 19–28% at 28, 90, 180 and 365 days. Scores computed at 48 hours and at 3–7 and 8–15 days after diagnosis predicted 28-day mortality significantly better than at diagnosis.
Comparison against the APASL framework — Choudhury 2017
2017 · n = 1,4021,402 patients from the APASL ACLF Research Consortium, used to compare the AARC score against CLIF-SOFA and MELD for predicting mortality in a population defined by the APASL rather than the EASL-CLIF criteria.
The AARC score achieved an AUROC of 0.80 in derivation and 0.78 in validation and was reported as superior to MELD and CLIF-SOFA in that cohort — a result that reflects the different populations the two frameworks define as much as the scores themselves.
Guideline adoption — EASL 2018
2018EASL clinical practice guidelines for the management of patients with decompensated cirrhosis, incorporating the CANONIC definition and grading of acute-on-chronic liver failure.
Adopted the EASL-CLIF organ failure criteria and ACLF grading as the European standard, and endorsed serial reassessment rather than reliance on the admission grade.
How it compares
CLIF-SOFA vs CLIF-C OF and CLIF-C ACLF
CLIF-C OF replaced CLIF-SOFA for diagnosis and CLIF-C ACLF is the better prognostic score — CLIF-SOFA is the original instrument, not the current best one.
Within a year of CANONIC the same consortium simplified CLIF-SOFA into the CLIF-C organ failure score for diagnosing the syndrome, and derived a separate prognostic score, CLIF-C ACLF, in 275 patients with ACLF. CLIF-C ACLF showed significantly higher predictive accuracy than MELD, MELD-Na and Child-Pugh, reducing prediction error by 19–28% at 28, 90, 180 and 365 days. If the question is prognosis, CLIF-C ACLF is the instrument the evidence supports. CLIF-SOFA retains value because it is the score the syndrome was originally defined with and the one much of the subsequent literature is expressed in, so understanding it is necessary to read the field.
CLIF-SOFA vs AARC-ACLF score (APASL)
They grade different syndromes rather than competing on the same one — APASL excludes previously decompensated patients, EASL-CLIF includes them, so the two frameworks are not scoring the same population.
This is the most consequential distinction in the ACLF literature and it is frequently glossed over. The APASL definition requires an acute hepatic insult producing jaundice and coagulopathy in a patient with chronic liver disease, and explicitly excludes those with prior decompensation; EASL-CLIF starts from acute decompensation of cirrhosis and defines the syndrome by organ failure and mortality. A patient with prior ascites who presents with renal failure has ACLF by EASL-CLIF and falls outside APASL entirely. In the APASL cohort of 1,402 patients the AARC score outperformed CLIF-SOFA with an AUROC of 0.80, which is a real finding about that population rather than evidence that one score is universally better. Use whichever framework matches your population, and say which one you used.
CLIF-SOFA vs MELD-Na
MELD-Na ranks waiting-list priority from four laboratory values; CLIF-SOFA identifies a syndrome across six organ systems and predicts short-term mortality better in that population.
MELD-Na was fitted to three-month mortality in a waiting-list population and has no representation of circulatory or respiratory failure, encephalopathy or vasopressor dependence. In acute-on-chronic liver failure those are precisely the variables driving outcome, which is why the CLIF-C ACLF score outperformed MELD and MELD-Na in the CANONIC dataset. The two do different jobs: MELD-Na remains the allocation instrument and should still be calculated, while CLIF-SOFA determines whether the patient is in a syndrome whose mortality MELD-Na will understate.
CLIF-SOFA vs Child-Pugh score
Child-Pugh grades chronic hepatic function and is not built for an acute deterioration — CLIF-C ACLF reduced prediction error by 19–28% against it in ACLF.
Child-Pugh describes where a patient sits in the natural history of their cirrhosis, using variables that move slowly, and it has no way to represent renal, circulatory or respiratory failure. A Child-Pugh B patient can be in grade 3 ACLF with a 77% 28-day mortality, and the Child-Pugh grade will not have changed. It remains useful as the baseline against which an acute deterioration is measured, and it is still the standard in some non-transplant contexts, but it should not be the instrument on which acute prognosis rests in this setting.
Pearls & pitfalls
- The kidney fails at a subscore of 2, not 3. Applying a uniform threshold of 3 across all six systems under-diagnoses ACLF, and it does so in the system where failure is both commonest and most prognostically loaded.
- One organ failure does not automatically mean grade 1 ACLF. A single liver, coagulation, circulatory or respiratory failure only qualifies when accompanied by creatinine 1.5–1.9 mg/dL and/or hepatic encephalopathy grade I–II; single kidney failure qualifies on its own; single cerebral failure needs creatinine 1.5–1.9. A patient with one non-renal organ failure and normal renal and cerebral function has organ failure without ACLF.
- Renal dysfunction for that qualifying rule is creatinine 1.5–1.9 mg/dL specifically, which is narrower than the kidney subscore of 1 (1.2–1.99). Reading the subscore instead of the creatinine wrongly admits a creatinine of 1.3.
- The total is close to meaningless on its own. Report the number of organ failures and the grade; a CLIF-SOFA of 12 tells the next clinician nothing about which organs are failing.
- Vasopressor use is circulatory failure. Do not score circulation from the achieved blood pressure in a patient on noradrenaline — the pressure is a treatment effect.
- Mechanical ventilation is respiratory failure regardless of oxygenation. An intubated patient with a PaO₂/FiO₂ of 350 still scores 3.
- The SpO₂/FiO₂ cut-offs differ from the PaO₂/FiO₂ cut-offs. Using 400/300/200/100 with a saturation-derived ratio will misclassify the respiratory subscore.
- An error in West Haven grading becomes an error in ACLF grade, because the encephalopathy grade is used directly as the cerebral subscore.
- Regrade daily. Trajectory at 48 hours and at three to seven days predicts 28-day mortality better than the admission value, and 112 of the 1,343 CANONIC patients developed ACLF only after enrolment.
- About 40% have no identified precipitant. That is a known feature of the syndrome, not evidence that the search was adequate — check for infection, alcohol, bleeding and drugs explicitly.
- APASL defines ACLF differently and excludes previously decompensated patients. A patient can have ACLF by EASL-CLIF and not by APASL, and the literature does not always say which framework it is using.
Critical actions
- Count organ failures using the correct per-system thresholds, remembering that the renal threshold is a creatinine of 2.0 mg/dL.
- Search actively for a precipitant: diagnostic paracentesis where ascites is present, blood and urine cultures, a review of alcohol intake, assessment for variceal bleeding, and a drug history including herbal and over-the-counter agents.
- Recalculate at 48 hours and again at three to seven days, and record the trend — the trajectory carries more prognostic information than the admission grade.
- Assess transplant candidacy at grade 1, not at grade 3. The assessment takes time the patient may not have later.
- Escalate the care setting to match the grade rather than the current observations, since deterioration in this syndrome is rapid and often not heralded.
- Grade encephalopathy carefully with West Haven criteria, since that grade enters the score directly.
- State which framework you are using — EASL-CLIF or APASL — when documenting ACLF, because the two define different populations.
- Where transplantation is not feasible at grade 3, involve palliative care alongside active treatment rather than sequentially.
Why this score exists
The CANONIC investigators set out to answer a question that sounds definitional but is empirical: does acute-on-chronic liver failure exist as a distinct entity, or is it just severe decompensation. Their method was to refuse to define it by consensus. They enrolled consecutively, recorded organ failures prospectively, and let mortality draw the boundary — the 15% threshold at 28 days is the line they chose, and the combinations of organ failure that crossed it became the definition. That is why the resulting criteria contain an asymmetry no committee would have designed: the kidney fails at a lower subscore than every other system, because in this population that is what the data showed. The same logic explains the emphasis on serial assessment that the follow-up work made explicit — if the syndrome is defined by mortality risk, and risk changes over the first week, then a single measurement was never going to be the right unit of observation.
About the creator
First author, CANONIC study
Led the CANONIC study that defined acute-on-chronic liver failure and its grading from CLIF-SOFA organ failures.
Co-author of CANONIC; first author of the CLIF-C OF and CLIF-C ACLF development
Developed the simplified CLIF-C OF score for diagnosis and the CLIF-C ACLF prognostic score, and established that serial assessment outperforms the value at diagnosis.
Senior author, CANONIC; EASL-CLIF Consortium
Senior author of the CANONIC study and a principal architect of the EASL-CLIF Consortium's work on the syndrome.
Limitations
- The grade 1 definition is intricate — three qualifying patterns rather than a count — and is frequently applied as "any single organ failure" in practice and in secondary sources, which over-diagnoses ACLF in patients with isolated non-renal organ failure.
- CLIF-SOFA has been superseded for diagnosis by CLIF-C OF and for prognosis by CLIF-C ACLF, both from the same consortium.
- Derived entirely in European liver units, where alcohol-related cirrhosis predominates. Performance differs in populations dominated by hepatitis B, which is part of why the APASL framework exists.
- The grade at diagnosis is the weakest measurement available — the same score at 48 hours and at three to seven days predicts 28-day mortality significantly better.
- Requires arterial blood gas or oxygen saturation data and a mean arterial pressure, so it cannot always be completed at the point of admission.
- The cerebral subscore inherits the reproducibility problems of West Haven grading, which is weakest at the lower grades.
- Says nothing about the precipitant, and about 40% of patients have none identified — the score describes severity, not what to treat.
- Not an allocation instrument, and its use to argue transplant priority conflates prognosis with the separate policy question of how organs are distributed.
- The vasopressor and ventilation overrides mean the score partly measures treatment received, so it is sensitive to local thresholds for escalating support.
If you are the patient
CLIF-SOFA is a way of measuring how many of the body's major systems are failing in someone with long-standing liver disease who has become acutely unwell. Six are checked: the liver itself, the kidneys, blood clotting, the circulation, the brain, and the lungs. What matters most is not the total but how many of those six have crossed into failure, because that count identifies a condition called acute-on-chronic liver failure and indicates how urgent the situation is. In the large European study that defined it, people admitted with liver decompensation but no organ failure had around a 2% risk of dying within a month, while those with three or more failing organs had around a 77% risk. Two things follow from that, and both are hopeful rather than otherwise. First, the score is repeated over the following days rather than being decided on admission, because it can improve as well as worsen, and how it changes over the first week says more than the first reading. Second, there is almost always a trigger — an infection, a bleed, alcohol, or a new medication — and finding and treating it is the main thing that changes the outcome. This is also the point at which the team will consider whether a liver transplant is an option, and they will do that assessment early rather than waiting, because it takes time to arrange.
Frequently asked questions
What is CLIF-SOFA?#
An adaptation of the intensive care SOFA score to cirrhosis, scoring six organ systems — liver, kidney, coagulation, circulation, brain and lungs — from 0 to 4 each. Its purpose is to identify organ failures, and the count of failing organs defines acute-on-chronic liver failure and its grade under the EASL-CLIF framework established by the CANONIC study.
What are the ACLF grades and their mortality?#
In the CANONIC study of 1,343 patients, 28-day mortality was 22.1% at grade 1, 32.0% at grade 2 and 76.7% at grade 3, against 1.9% in patients admitted with decompensated cirrhosis who never developed organ failure. The grades correspond to one, two, and three or more organ failures respectively, with the important detail that the published grade 1 definition requires either kidney failure or another single organ failure accompanied by renal or cerebral dysfunction.
What counts as an organ failure in CLIF-SOFA?#
Liver at a bilirubin of 12 mg/dL or above, coagulation at an INR of 2.5 or above, circulation at vasopressor requirement, brain at West Haven grade III or IV, and lungs at a PaO₂/FiO₂ of 200 or below, an SpO₂/FiO₂ of 214 or below, or mechanical ventilation. The kidney is the exception: failure is defined at a creatinine of 2.0 mg/dL, a subscore of 2 rather than 3.
Why is the kidney threshold lower than the others?#
Because the criteria were derived empirically rather than by consensus. The CANONIC investigators identified which patterns of organ failure carried a 28-day mortality above 15%, and renal dysfunction crossed that line at a lower level of derangement than the other systems. It reflects how much prognostic weight kidney function carries in cirrhosis, and it is why applying a uniform threshold of 3 across all six systems under-diagnoses the syndrome.
Should CLIF-SOFA be repeated or calculated once?#
Repeated. The follow-up work from the same consortium showed that scores computed at 48 hours and at three to seven and eight to fifteen days after diagnosis predicted 28-day mortality significantly better than the value at diagnosis. In CANONIC itself, 112 of 1,343 patients developed ACLF only after enrolment, and their mortality approached that of patients who presented with it — none would have been identified by an admission assessment alone.
What is the difference between CLIF-SOFA and CLIF-C OF?#
CLIF-C OF is a simplified organ function score developed by the same consortium a year after CANONIC, and it replaced CLIF-SOFA for diagnosing ACLF. A separate score, CLIF-C ACLF, was derived for prognosis and outperformed MELD, MELD-Na and Child-Pugh, reducing prediction error by 19–28%. CLIF-SOFA remains worth understanding because it is the instrument the syndrome was originally defined with and the one much of the literature uses.
How does EASL-CLIF differ from the APASL definition of ACLF?#
Fundamentally. EASL-CLIF starts from acute decompensation of cirrhosis and defines the syndrome by organ failure and 28-day mortality, including patients with prior decompensation. APASL requires an acute hepatic insult causing jaundice and coagulopathy and explicitly excludes previously decompensated patients. A patient with established ascites who presents in renal failure has ACLF by EASL-CLIF and falls outside APASL altogether, so the two frameworks describe overlapping but genuinely different populations.
Does CLIF-SOFA determine transplant priority?#
No — allocation runs on MELD 3.0 or an equivalent national model, and CLIF-SOFA has no formal role in it. What the grade should drive is the timing of assessment: candidacy is best evaluated at grade 1, because the assessment takes time and the practical window narrows as further organs fail. A grade 3 patient who was never assessed earlier has usually run out of time rather than been declined.
How often is a precipitant found in ACLF?#
In roughly 60% of cases. About two in five patients have no identifiable trigger, which is a recognised feature of the syndrome rather than a sign of an incomplete work-up — though it should not stop the search. Bacterial infection, alcohol, variceal bleeding and drug-induced injury account for most of those that are identified, and all are actionable.
References
Validation and evidence
- Jalan R, Saliba F, Pavesi M, Amoros A, Moreau R, Gines P, Levesque E, Durand F, Angeli P, Caraceni P, Hopf C, Alessandria C, Rodriguez E, Solis-Muñoz P, Laleman W, Trebicka J, Zeuzem S, Gustot T, Mookerjee R, Elkrief L, Soriano G, Cordoba J, Morando F, Gerbes A, Agarwal B, Samuel D, Bernardi M, Arroyo V. Development and validation of a prognostic score to predict mortality in patients with acute-on-chronic liver failure. J Hepatol. 2014;61(5):1038-1047 (CLIF-C OF and CLIF-C ACLF).
- Choudhury A, Jindal A, Maiwall R, et al. Liver failure determines the outcome in patients of acute-on-chronic liver failure (ACLF): comparison of APASL ACLF research consortium (AARC) and CLIF-SOFA models. Hepatol Int. 2017;11(5):461-471.