Daily albumin infusion fails to reduce infection, renal dysfunction in decompensated cirrhosis

Daily albumin infusion to achieve serum albumin levels of ≥30 g/L did not reduce the risk of infection, renal dysfunction, or mortality in individuals with acute decompensated cirrhosis, according to results of the ATTIRE* study presented at ILC 2020.

This multicentre (35 hospitals in the UK), open-label trial involved 829 patients (mean age 53.7 years, 70 percent male) hospitalized with acute decompensated cirrhosis and serum albumin levels <35 g/L. Within 72 hours of admission, they were randomized 1:1 to receive human albumin solution (HAS) 20% daily** or standard care (control) for <14 days, discharge from hospital, or death.

At baseline, median creatinine was 68 μmol/L, bilirubin 94 μmol/L, MELD*** score 19.6, and albumin 23 g/L, with no significant difference between groups. Alcohol misuse was reported in 79 percent of patients, 28 percent had infection, and 53 percent were on antibiotics. About 29 percent of patients had baseline serum albumin levels <20 g/L. Aetiology of cirrhosis was primarily alcohol-related and the main reason for hospitalization for decompensation was new or worsening ascites.

Patients with advanced hepatocellular carcinoma with a life expectancy <8 weeks, those receiving only palliative care, and those with known or suspected severe cardiac dysfunction were excluded.

Patients in the intervention and control arms received a median HAS volume of 1,000 and 100 mL, respectively (p<0.0001). The intervention led to serum albumin of >30 g/L by day 3 which was sustained throughout the treatment period, with a mean difference of 4–7 g/L between groups from day 2.

The primary composite endpoint of new infection, renal dysfunction, or mortality between days 3 and 15 of treatment occurred in a comparable proportion of patients in the intervention and control groups (30.2 percent vs 30.9 percent; odds ratio [OR], 0.968, 95 percent confidence interval [CI], 0.716–1.307; p=0.830). [ILC 2020, abstract LBO02]

Subgroup analysis showed that the results were consistent across multiple subgroups including baseline serum albumin levels, number of dysfunctional organs, receipt of antibiotics, age, and sex.

There was no difference between the intervention and control groups when components of the primary endpoint were assessed separately (new infection: 21.0 percent vs 18.4 percent; OR, 1.196, 95 percent CI, 0.845–1.693; p=0.313; renal dysfunction: 10.9 percent vs 15.0 percent; OR, 0.674, 95 percent CI, 0.445–1.022; p=0.063; mortality: 7.7 percent vs 8.2 percent; OR, 0.936, 95 percent CI, 0.566–1.550; p=0.798).

There was also no between-group difference for mortality at 28 days (13.5 percent vs 15.7 percent; OR, 0.834, 95 percent CI, 0.559–1.245; p=0.374) or 3 months (20.3 percent vs 19.3 percent; OR, 1.084, 95 percent CI, 0.761–1.544; p=0.6548).

Grade 3 (severe) and 4 (life-threatening) adverse events (AEs) occurred more often in the intervention vs control group (n=47 vs 20 and 32 vs 20, respectively). There were 43 and 47 deaths in the intervention and control group, respectively. The most common serious AEs were respiratory tract infection in both groups, and pulmonary oedema/fluid overload in the intervention and gastrointestinal haemorrhage in the control group.

“Patients with acute decompensated cirrhosis are at high risk of infection,” presented study investigator Dr Louise China from University College London, UK. “There is growing experimental evidence that albumin has anti-inflammatory … effects … and that a low serum albumin is associated with increased mortality from nosocomial infection.”

“[The ATTIRE trial] demonstrated there was no benefit in administering targeted intravenous albumin vs standard care in hospitalized patients with acute decompensated cirrhosis. This study advocates a re-evaluation of use of albumin in advanced liver disease,” she concluded.