Hepatic recompensation has emerged as a distinct stage of cirrhosis 1, supporting the concept of disease regression even in the stage of decompensated cirrhosis once the underlying aetiological driver is controlled or removed 2. Although previous studies have consistently demonstrated improved outcomes following recompensation 3-6, baseline predictors of a patient's recompensation likelihood vary across cohorts, aetiologies, and clinical scenarios. Although liver disease severity has consistently been identified as a key determinant of recompensation 3-5, the predictive value of hepatic function markers appears context-dependent. For instance, while lower MELD robustly predicted recompensation in clinically stable patients 3, 5, higher MELD was linked to increased recompensation rates in patients with ongoing hepatic necroinflammation due to viral hepatitis 6. Xia et al. 7 now provide insights into recompensation in decompensated hepatitis B virus (HBV) cirrhosis, reporting a recompensation rate of 78% and providing data on the context-specific role of laboratory markers of hepatic function and portal hypertension. They also introduce the term ‘unstable recompensation’ to describe patients who either experienced another decompensating event after recompensation or achieved recompensation only after multiple decompensating events. The ‘unstable recompensated’ patients comprised 31% of the recompensated cohort and had an intermediate prognosis compared to those with stable recompensation or ongoing decompensation. Patients who achieved stable recompensation not only exhibited more pronounced hepatic inflammation (higher transaminases) at baseline, but also a higher baseline MELD and Child-Pugh score—primarily driven by elevated bilirubin levels. Simultaneously, those with stable recompensation also showed the largest reductions in bilirubin, MELD, and Child-Pugh score at 6 months, thus supporting the hypothesis that elevated baseline transaminases and bilirubin seem to reflect a severe and acute but potentially reversible necroinflammatory process rather than chronic (and thus, less reversible) hepatic functional impairment. Moreover, despite similar baseline albumin levels, albumin increased most substantially in patients with stable recompensation. Accordingly, while there was no link between baseline albumin and the likelihood of recompensation, albumin re-assessed at 6 months was a predictor of recompensation. These findings reinforce the proposed value of monitoring dynamic changes in hepatic function to predict recompensation, which is consistent with the association of on-treatment albumin levels in HBV 6, 8 and changes in MELD and albumin after 90 days of abstinence in alcohol-related cirrhosis 9. Conversely, baseline platelet count was higher in patients with stable recompensation, which supports the impact of portal hypertension severity on the likelihood of recompensation 3, 6. In contrast to MELD and albumin, changes in platelet count in repeated measurements did not differ significantly between patients with stable vs. unstable recompensation and versus those with ongoing decompensation, aligning with evidence that thrombocytopenia may persist after recompensation, even if clinically significant portal hypertension resolves 10. Overall, we would like to congratulate Xia et al. for their innovative approach of assessing prognostic links of dynamics in laboratory markers of hepatic function in patients achieving recompensation. Since hepatic function at presentation with decompensated cirrhosis may be (temporarily) influenced by necroinflammation, the prognostic value of lab-based assessments may be more informative when re-evaluated after a period of clinical stabilisation under aetiologic therapy. Benedikt Silvester Hofer: conceptualization, writing – original draft, writing – review and editing. Jan Embacher: conceptualization, writing – review and editing. Georg Semmler: conceptualization, writing – review and editing. Thomas Reiberger: conceptualization, writing – review and editing, supervision. The authors have nothing to report. B.S.H. received travel support from Ipsen and Falk. J.E. has nothing to disclose. G.S. received speaking honoraria from DiaSorin. T.R. received grant support from Abbvie, Boehringer Ingelheim, Gilead, Intercept/Advanz Pharma, MSD, Myr Pharmaceuticals, Philips Healthcare, Pliant, Siemens, and W. L. Gore and travel support from Abbvie, Boehringer Ingelheim, Dr. Falk Pharma, Gilead, and Roche. This article is linked to Xia et al. papers. To view these articles, visit https://doi.org/10.1111/apt.70430 and https://doi.org/10.1111/apt.70542. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Hofer et al. (Tue,) studied this question.