Real-world experience with long-term albumin in patients with cirrhosis and ascites

Abstract

Background & aims: Long-term albumin (LTA) is currently standard of care for patients with decompensated cirrhosis in many Italian hepatology centres. In this real-life study, we aimed to describe patient, logistical and treatment-related characteristics in daily clinical practice and to identify predictors of response.

Methods: We performed a multicentre, retrospective, observational study in patients with cirrhosis and ascites receiving LTA between 01/2016 and 02/2022 and followed until death, TIPS (transjugular intrahepatic portosystemic shunt) placement, transplantation or 02/2023.

Results: A total of 312 patients, the majority with alcohol-related cirrhosis, were included. At baseline, median Child-Pugh, MELD, and MELD-Na were 8, 15, and 18, respectively. Ascites was grade 2 in 55% of patients, grade 3 in 35% and refractory in 27%, while 47% had received large volume paracentesis in the previous 6 months. Median LTA was 10 months with a median dose of 40 g/week. Ascites resolved to grade 0-1 in 34% of patients within the first 3 months and 56% by the end of treatment. Predictors of ascites resolution were age (p = 0.007), baseline grade of ascites (p = 0.007), no paracentesis in the previous 6 months (p = 0.001), aetiological treatment in the past 12 months or during LTA (p = 0.005), weekly albumin dose (p = 0.014) and serum albumin concentration of 40 g/L after 1 month of treatment (p = 0.017). Of the 83 patients with refractory ascites at inclusion, 26% had grade 0/1 ascites at the last observation. No severe albumin-related side-effects were reported and only 1% discontinued for logistical reasons.

Conclusions: LTA is feasible as an outpatient treatment for the management of ascites. In the current study, ascites resolved in more than half of patients receiving LTA on top of diuretics, including in some with refractory ascites. Predictors of response to LTA provide useful information for tailoring treatment.

Impact and implications: The ANSWER randomised-controlled trial has shown that long-term albumin treatment (LTA) is an effective approach for the management of patients with cirrhosis and ascites. This observational study provides novel information on target patients, modalities and length of treatment, predictors of ascites resolution, stopping criteria, and clinical trajectories of patients on treatment. LTA is a feasible option in the daily clinical practice for the management of ascites when given on top of diuretics. Rather than an alternative therapy, LTA should be integrated with the other treatment options already available for patients with difficult-to-treat ascites. The predictive factors of response identified in the present study can help physicians to individualise LTA and optimise the decision-making process.

Keywords: ANSWER; TIPS; ascites; decompensated cirrhosis; human albumin; liver cirrhosis; outpatient; paracentesis; portal hypertension; refractory ascites.

Graphical abstract

Fig. 1

Patients included in the study and reasons for LTA treatment interruption. Median treatment duration was 9 (5-16) months in patients who stopped LTA due to clinical improvement, 10 (6-12) months in patients who received LT, 6 (3-13) months in patients who received TIPS, 6 (3-11) months in patients who stopped for other reasons and 11 (5-16) months in patients who died during LTA. LT, liver transplantation; LTA, long-term albumin; TIPS, transjugular intrahepatic portosystemic shunt.

Fig. 2

Sankey plots showing the patients with grade 1, 2 and 3 ascites at inclusion and those with grade 0-1, 2 and 3 ascites at the last observation before the end of treatment for any reason or at the last visit for those still on treatment on February 28^th, 2023. (A) The whole study population (N = 312) and (B) the subgroup of patients who did not receive any aetiological treatment for cirrhosis during long-term albumin treatment or in the year prior to enrolment (n = 198).

Fig. 3

Independent predictors of ascites resolution. (A) Independent predictors of ascites resolution at the end of treatment or at last available visit in patients still on treatment at the end of follow-up. (B) Independent predictors of ascites resolution within 3 months from study inclusion. Odds ratio and 95% CI were estimated from multivariable logistic regression performed on 100 imputation datasets. Values from continuous variables were divided by the IQR except for albumin dose which was divided by the median.

Fig. 4

Percentage of patients with ascites grade 0/1 at 3 months, 6 months, and 12 months after initiation of long-term albumin treatment.

Fig. 5

Cumulative incidence of treatment discontinuation due to clinical improvement up to 18 months. Censored patients are those who discontinued treatment for other reasons (death, LT, TIPS, other causes) or who interrupted follow-up before 18 months. The cumulative incidence was estimated according to the Kaplan-Meier method. LT, liver transplantation; TIPS, transjugular intrahepatic portosystemic shunt.