Skip to content

Cirrhosis and decompensation

TL;DR — Compensated advanced chronic liver disease (cACLD) due to MASLD is a slow state that becomes fast at a single inflection point. In an international 17-centre cohort of 6,061 patients with MASLD-cACLD, the 5-year cumulative incidence of first decompensation was only 3.5% (95% CI 3.0–4.1), but that event raised the cause-specific hazard of liver-related death 18.9-fold (10.8–32.9); among those who decompensated once, further decompensation reached 43.9% (37.2–50.2) at 5 years and added a further 1.52-fold hazard (Pennisi 2025, PMID 40550340). What separates the slow from the fast state is clinically significant portal hypertension (CSPH, HVPG ≥10 mmHg): with CSPH, 5-year decompensation incidence was 30.7% and liver-related mortality 21.4%; without it, 9.4% and 0.8% — and variceal bleeding did not occur at all without CSPH (Paternostro 2024, PMID 38823501). Ascites is the first event in about 70% of cases, and median survival from first decompensation to death or transplant is 2.0 years (Noureddin 2024, PMID 38571305). MASLD cirrhosis is systematically disadvantaged on the transplant waiting list: listed at lower MELD despite greater portal hypertension, transplanted less often (HR 0.867 at 1 year) and dying on the list more often (HR 1.25 at 1 year) than non-NASH cirrhosis (Lim 2023, PMID 37307997). Every approved MASH drug stops at F3.

The two states of cACLD

Portal hypertension, not fibrosis stage, is what divides them. European multicentre cohort of 340 patients with MASLD-cACLD characterised by hepatic venous pressure gradient (56.2% male, median age 62, median MELD 8, 71.2% with diabetes; CSPH present in 209 [61.5%]), median follow-up 41.5 months (Paternostro 2024, PMID 38823501):

Outcome CSPH (HVPG ≥10 mmHg) No CSPH
Hepatic decompensation, 2 years 10.0% 2.4%
Hepatic decompensation, 5 years 30.7% 9.4%
Liver-related mortality, 2 years 3.3% —
Liver-related mortality, 5 years 21.4% 0.8%
Variceal bleeding occurs never occurred

CSPH carried a subdistribution hazard ratio of 5.13 (p<0.001) for decompensation, and HVPG remained an independent continuous predictor after adjustment — adjusted SHR 1.12 per mmHg for decompensation (p<0.001) and 1.20 per mmHg for liver-related death (p<0.001). The practical implication is that HVPG (or a validated non-invasive surrogate) stratifies MASLD-cACLD far more sharply than F3-versus-F4 does.

First and further decompensation

International retrospective cohort, 17 centres, 6,061 consecutive patients with cACLD due to MASLD diagnosed clinically (liver stiffness >10 kPa) or histologically (F3–F4), with events defined by Baveno VII criteria and a multistate model (Pennisi 2025, PMID 40550340):

Transition 5-year cumulative incidence Effect on liver-related death
Compensated → first decompensation 3.5% (3.0–4.1) cause-specific HR 18.9 (10.8–32.9)
First → further decompensation 43.9% (37.2–50.2) additional HR 1.52 (1.02–2.34)
HCC, overall cohort — HR 2.95 (2.02–4.31)
HCC, after first decompensation — HR 1.43 (1.03–2.00)

Ascites was the most common and most impactful first event, ahead of variceal bleeding; acute and non-acute decompensation contributed similarly to liver-related mortality. The authors also record that extrahepatic deaths represented a considerable burden, reflecting the metabolic risk of the underlying disease — see cardiovascular and extrahepatic outcomes.

The sequence and its timing come from a separate 2,016-patient MASLD cohort in which 220 (11%) decompensated over a median 3.2 years: ascites first in 69.5%, hepatic encephalopathy in 25%, variceal haemorrhage in 13.6%; second event at a median 0.7 years, third at 1.3 years after the second; median survival from first decompensation to death or transplant 2.0 years; 33.1% died and 14% were transplanted (Noureddin 2024, PMID 38571305).

Predicting events better than histology can

Fibrosis stage is the regulatory reference for trial entry, and it is outperformed. In 699 biopsy-proven F3–F4 MASLD patients (56 liver-related events, 8.0%, of which 51 occurred in F4), the non-invasive ANTICIPATE-NASH models achieved a C statistic of 0.93 for liver-related events versus 0.67 for histology; adding histology did not improve prediction, and thresholds were identified above which F3 patients developed events and below which F4 patients did not. Validation in 1,396 F3–F4 patients from four clinical trials using the regulatory endpoint reproduced the gap (0.84 vs 0.64) (Aceituno 2026, PMID 41212130).

This matters for two reasons: it argues that trials selecting on histological stage are enrolling a mis-stratified population, and it means an F3 patient above the threshold may carry more event risk than an F4 patient below it — which the F3/F4 licensing boundary ignores. See noninvasive assessment and histology and biopsy.

Serial liver stiffness carries similar information: patients with persistently high Agile 3+ scores had 30.1 liver-related events per 1,000 person-years versus 0.6 with persistently low scores (Lin 2024, PMID 38512249).

Detecting portal hypertension without a catheter — and where it fails in this disease

Because CSPH is the variable that separates the slow from the fast state, non-invasive detection of it is the practical problem. The Baveno VII criteria — rule in at liver stiffness ≥25 kPa, rule out at LSM ≤15 kPa with platelets ≥150 ×10⁹/L — work across aetiologies except this one. An individual-patient-data meta-analysis of five validation studies (1,433 patients with cACLD, CSPH prevalence 34–62%, HVPG as reference) found (Bañares 2026, PMID 41138818):

Criterion Performance
Rule-in, LSM ≥25 kPa, pooled PPV 92% across aetiologies — but not in MASLD with obesity, where PPV was 0.67
Rule-out, LSM ≤15 kPa + platelets ≥150 pooled NPV 99%, all aetiologies
ANTICIPATE ± NASH ≥75% risk threshold PPV 95% overall; in MASLD with obesity PPV rose 0.67 → 0.83, p<0.001

The failure is specific to the MASLD-with-obesity subgroup in this analysis. The ANTICIPATE ± NASH calculation uses BMI, liver stiffness and platelet count and improved, but did not fully restore, rule-in PPV in that subgroup (0.67 to 0.83).

Spleen stiffness closes most of the remaining grey zone. Measured by vibration-controlled transient elastography with the 100 Hz spleen-specific module, SSM correlated with HVPG at r=0.74 (p<0.0001) in 60 MAFLD patients, and dual cut-offs of <40.9 and >49.9 kPa gave an AUC of 0.95 for CSPH; adding these to the Baveno VII criteria sequentially or in combination cut the indeterminate zone from 60% to 15–20% while keeping NPV and PPV above 90% (Odriozola 2023, PMID 36912787). In a larger 16-centre European derivation and temporal-validation study (407 patients with Child-Pugh A cACLD, 40% MASLD, 59% with CSPH), SSM alone matched LSM (AUC 0.779 vs 0.781 derivation; 0.830 vs 0.804 validation), and the four-variable NICER model (SSM + LSM + platelets + BMI) significantly outperformed ANTICIPATE ± NASH (AUC 0.889 vs 0.849, p=0.022 derivation; 0.906 vs 0.863, p=0.012 validation) (Jachs 2024, PMID 39326431). Two caveats belong on the record: the Odriozola cohort is 85 patients from one centre, and the Jachs study was funded by the manufacturer of the measurement device.

Burnt-out cirrhosis and attribution

As MASLD reaches cirrhosis, hepatic steatosis can disappear, so the aetiological signature is lost exactly where the disease burden is greatest; guidance on histological reporting addresses the special situation of MASH in advanced fibrosis or cirrhosis for this reason (Brunt 2021, PMID 33111374). Longitudinal data confirm the phenomenon prospectively — liver fat loss in advanced MASLD has been described as moving from "burnt-out" to "burning-out" (Liu 2025, PMID 40113099). The consequence is that MASLD-attributable cirrhosis is systematically undercounted and some of it is recorded as cryptogenic.

Transplantation

MASLD/NASH cirrhosis is among the fastest-growing indications for liver transplantation, and the waitlist data show a specific structural disadvantage. Scientific Registry of Transplant Recipients, patients listed 1 January 2016 – 31 December 2021, NASH cirrhosis (n=8,120) versus non-NASH cirrhosis (n=21,409) (Lim 2023, PMID 37307997):

Metric NASH vs non-NASH
MELD at listing lower, despite a greater burden of portal hypertension, especially at lower MELD
Transplant probability, 90 days HR 0.873 (p<0.001)
Transplant probability, 1 year HR 0.867 (p<0.001)
Transplant probability, MELD >30 HR 0.705 at 90 days; 0.672 at 1 year (both p<0.001)
Waitlist mortality, 90 days HR 1.15 (p<0.001)
Waitlist mortality, 1 year HR 1.25 (p<0.001)
Main driver of MELD rise leading to transplant serum creatinine (bilirubin in non-NASH)

The mechanism the authors propose is that MELD, built on bilirubin, INR and creatinine, under-captures the mortality risk of a disease whose portal hypertension develops out of proportion to synthetic dysfunction, and whose renal component (see cardiovascular and extrahepatic outcomes) is the element that eventually moves the score. The differences were most pronounced at lower MELD scores at listing.

Projected demand is rising steeply: a US agent-based model forecasts MASLD-attributable liver transplants increasing from 1,717 per year in 2020–2025 to 6,720 per year in 2046–2050, with liver-related deaths rising from 30,500 to 95,300 annually (Le 2025, PMID 39821400).

Treatment stops at F3

Intervention Evidence in cirrhosis
Resmetirom Not approved in cirrhosis; MAESTRO-NASH enrolled F1B–F3, and expert guidance's principal instruction is to exclude cirrhotic patients (Noureddin 2024, PMID 39038768)
Semaglutide Not approved in MASH cirrhosis; phase 2 in compensated NASH cirrhosis showed fibrosis improvement in 11% vs 29% placebo (OR 0.28, 95% CI 0.06–1.24, p=0.087) (Loomba 2023, PMID 36934740)
Efruxifermin SYMMETRY negative at week 36 in compensated MASH cirrhosis (19% vs 13%, p=0.52); week 96 difference 16 points (95% CI 2–30) (Noureddin 2025, PMID 40341827)
Selonsertib STELLAR-4 negative in compensated cirrhosis (14%, 13%, 13% placebo) (Harrison 2020, PMID 32147362)
Bariatric surgery Safety in cirrhosis explicitly unresolved (Lefere 2021, PMID 34002452)
GLP-1 RA, observational In 1,452 VHA patients with MASLD, diabetes and cirrhosis at baseline, GLP-1 RA versus DPP-4i showed no association with any outcome; in 14,606 without cirrhosis, cirrhosis progression HR 0.86 (0.75–0.98) and mortality HR 0.89 (0.81–0.98) (Kanwal 2024, PMID 39283612)
Vitamin E Observational: transplant-free survival 78% vs 49% and decompensation 37% vs 62% in bridging fibrosis or cirrhosis, adjusted HR 0.30 (0.12–0.74) (Vilar-Gomez 2020, PMID 30506586)

The Kanwal analysis makes the point most sharply — the authors conclude that the absence of a protective association in established cirrhosis "underscor[es] the importance of treatment earlier in the disease course." Whether cirrhosis is biologically irreversible or the trials are simply too short and small is unresolved; the efruxifermin week-36-negative/week-96-positive pattern favours the latter.

Recompensation has now been documented specifically in metabolic-dysfunction-related steatotic liver disease. In an observational cohort of 344 patients with decompensated SLD cirrhosis (158 MASLD and 186 MetALD), 64 (18.6%) met recompensation criteria; the proportions were 17.7% in MASLD and 19.4% in MetALD. Weight loss of at least 10% (subdistribution HR 7.42), glycaemic control (1.43) and PEth-corroborated alcohol abstinence (1.87) were associated with recompensation. These associations do not establish treatment causality, but they refute the premise that recompensation has not been observed in MASLD (Premkumar 2026, PMID 42467948).

Open questions

  • Can the MASLD-with-obesity rule-in failure be closed prospectively? The ANTICIPATE ± NASH threshold raises PPV from 0.67 to 0.83 in this subgroup and spleen stiffness cuts the grey zone from 60% to 15–20% (PMIDs: 41138818, 36912787, 39326431), but all of this is retrospective or modelling work, one key study is device-manufacturer funded, and no prospective outcome study has used a spleen-stiffness-guided pathway to select patients for beta-blocker prophylaxis.
  • Is MASH cirrhosis modifiable at all? Three cited randomised trials in compensated MASH cirrhosis failed their primary endpoint (PMIDs: 36934740, 40341827, 32147362), and one large observational analysis found no GLP-1 benefit once cirrhosis was established (PMID 39283612). Recompensation is nevertheless observed in 17.7% of an MASLD subgroup and is associated with weight loss and glycaemic control (PMID 42467948). Query run 2026-09-02: (MASLD OR NAFLD OR NASH OR MASH) AND ("compensated advanced chronic liver disease" OR cACLD OR "clinically significant portal hypertension" OR "Baveno") — 138 records; no completed positive phase 3 trial in cirrhosis was identified. A recruiting randomised trial now tests TIPS plus sleeve gastrectomy against medical/lifestyle care in 70 patients with cirrhosis, portal hypertension and severe obesity (NCT07058155); it has not yet reported outcomes.
  • Should MELD be modified for MASLD? Waitlist data show listing at lower MELD with more portal hypertension, less transplantation and more waitlist death, with creatinine as the driver (PMID 37307997). A 44,037-candidate registry study, including 12,790 with MASH, found that MELD 3.0 up-categorised 65.4% of MASH candidates and down-categorised 11.2%; up-categorisation was associated with mortality (HR 1.14) but not significantly greater transplantation (HR 1.04). This retrospective post-implementation analysis did not show improved access; prospective evaluation remains absent (Lim 2026, PMID 41284513).
  • Should HVPG or a portal-hypertension surrogate replace fibrosis stage for trial entry? CSPH separates 5-year liver-related mortality by a factor of 26 (21.4% vs 0.8%) (PMID 38823501), and ANTICIPATE-NASH outperforms histology (C 0.93 vs 0.67) (PMID 41212130), yet trials enrol and drugs are licensed by histological stage.
  • How much MASLD cirrhosis is recorded as cryptogenic? Steatosis disappears at the cirrhotic stage (PMIDs: 33111374, 40113099), and no study has quantified the resulting misattribution.
  • Can recompensation be caused by metabolic treatment? MASLD-specific recompensation has been observed (17.7% in the MASLD subgroup), with weight loss and glycaemic control as strong associations (PMID 42467948). As of the 2026-09-02 audit, no randomised treatment study establishes that either intervention causes recompensation.

References

  1. Pennisi G, Di Maria G, Wong VW, et al. Impact of first and further decompensation in patients with compensated ACLD due to MASLD. J Hepatol. 2025;83(6):1266-1277. PMID 40550340
  2. Paternostro R, Kwanten WJ, Hofer BS, et al. Hepatic venous pressure gradient predicts risk of hepatic decompensation and liver-related mortality in patients with MASLD. J Hepatol. 2024;81(5):827-836. PMID 38823501
  3. Noureddin N, Huang DQ, Bettencourt R, et al. Natural history of clinical outcomes and hepatic decompensation in metabolic dysfunction-associated steatotic liver disease. Aliment Pharmacol Ther. 2024;59(12):1521-1526. PMID 38571305
  4. Aceituno L, Bañares J, Pons M, et al. The ANTICIPATE-NASH Models Stratify Better the Risk of Clinical Events Than Histology in Metabolic Dysfunction-Associated Steatotic Liver Disease Patients With Advanced Chronic Liver Disease. Gastroenterology. 2026;170(2):385-394. PMID 41212130
  5. Lim WH, Ng CH, Tan D, et al. Natural history of NASH cirrhosis in liver transplant waitlist registrants. J Hepatol. 2023;79(4):1015-1024. PMID 37307997
  6. Kanwal F, Kramer JR, Li L, et al. GLP-1 Receptor Agonists and Risk for Cirrhosis and Related Complications in Patients With Metabolic Dysfunction-Associated Steatotic Liver Disease. JAMA Intern Med. 2024;184(11):1314-1323. PMID 39283612
  7. Loomba R, Abdelmalek MF, Armstrong MJ, et al. Semaglutide 2·4 mg once weekly in patients with non-alcoholic steatohepatitis-related cirrhosis: a randomised, placebo-controlled phase 2 trial. Lancet Gastroenterol Hepatol. 2023;8(6):511-522. PMID 36934740
  8. Noureddin M, Rinella ME, Chalasani NP, et al. Efruxifermin in Compensated Liver Cirrhosis Caused by MASH. N Engl J Med. 2025;392(24):2413-2424. PMID 40341827
  9. Harrison SA, Wong VW, Okanoue T, et al. Selonsertib for patients with bridging fibrosis or compensated cirrhosis due to NASH: Results from randomized phase III STELLAR trials. J Hepatol. 2020;73(1):26-39. PMID 32147362
  10. Vilar-Gomez E, Vuppalanchi R, Gawrieh S, et al. Vitamin E Improves Transplant-Free Survival and Hepatic Decompensation Among Patients With Nonalcoholic Steatohepatitis and Advanced Fibrosis. Hepatology. 2020;71(2):495-509. PMID 30506586
  11. Brunt EM, Kleiner DE, Carpenter DH, et al. NAFLD: Reporting Histologic Findings in Clinical Practice. Hepatology. 2021;73(5):2028-2038. PMID 33111374
  12. Liu WY, Zheng KI, Byrne CD, et al. From "Burnt-Out" to "Burning-Out": Capturing Liver Fat Loss in Patients With Advanced Metabolic Dysfunction-Associated Steatotic Liver Disease From a Dynamic Perspective. Gastroenterology. 2025;169(2):326-336.e4. PMID 40113099
  13. Lin H, Lee HW, Yip TC, et al. Vibration-Controlled Transient Elastography Scores to Predict Liver-Related Events in Steatotic Liver Disease. JAMA. 2024;331(15):1287-1297. PMID 38512249
  14. Le P, Tatar M, Dasarathy S, et al. Estimated Burden of Metabolic Dysfunction-Associated Steatotic Liver Disease in US Adults, 2020 to 2050. JAMA Netw Open. 2025;8(1):e2454707. PMID 39821400
  15. Noureddin M, Charlton MR, Harrison SA, et al. Expert Panel Recommendations: Practical Clinical Applications for Initiating and Monitoring Resmetirom in Patients With MASH/NASH and Moderate to Noncirrhotic Advanced Fibrosis. Clin Gastroenterol Hepatol. 2024;22(12):2367-2377. PMID 39038768
  16. Lefere S, Onghena L, Vanlander A, et al. Bariatric surgery and the liver-Mechanisms, benefits, and risks. Obes Rev. 2021;22(9):e13294. PMID 34002452
  17. Premkumar M, Kulkarni AV, Sarin SK. Recompensation in decompensated cirrhosis. Lancet Gastroenterol Hepatol. 2025;10(11):1028-1040. PMID 40780235
  18. Premkumar M, et al. Recompensation of decompensated cirrhosis in a spectrum of metabolic dysfunction-related steatotic liver disease, with PEth-corroborated alcohol abstinence and modification of cardiometabolic risk factors. Hepatology. 2026. PMID 42467948
  19. Lim WH, Ng CH, et al. Impact of Model for End-Stage Liver Disease 3.0 on Waitlist Outcomes of Metabolic Dysfunction-Associated Steatohepatitis Cirrhosis Among Liver Transplant Candidates. Clin Gastroenterol Hepatol. 2026;24(8):2181-2190. PMID 41284513
  20. Bañares J, Pons M, Reiberger T, et al. Refining the Baveno VII criteria for clinically significant portal hypertension: An individual patient data meta-analysis. J Hepatol. 2026;84(3):543-553. PMID 41138818
  21. Odriozola A, Puente Á, Cuadrado A, et al. High accuracy of spleen stiffness measurement in diagnosing clinically significant portal hypertension in metabolic-associated fatty liver disease. Liver Int. 2023;43(7):1446-1457. PMID 36912787
  22. Jachs M, Odriozola A, Turon F, et al. Spleen stiffness measurement by vibration-controlled transient elastography at 100 Hz for non-invasive predicted diagnosis of clinically significant portal hypertension in patients with compensated advanced chronic liver disease: a modelling study. Lancet Gastroenterol Hepatol. 2024;9(12):1111-1120. PMID 39326431