Effect of standard dose paracetamol versus placebo as antipyretic therapy on liver injury in adult dengue infection: a multicentre randomised controlled trial

Summary: Background: Dengue is a common cause of acute liver failure in tropical countries. Paracetamol is the recommended antipyretic for dengue. Related observational studies in dengue have suggested that excessive paracetamol intake is related to hepatic injury. We aimed to evaluate whether standard dose paracetamol as an antipyretic in dengue infection caused transaminase elevation, and to evaluate the efficacy of paracetamol. Methods: In this randomised, double-blind, placebo-controlled trial, adult participants (aged ≥18 years) with dengue, as confirmed by either positive NS1 antigen, positive dengue IgM antigen with thrombocytopenia, or positive PCR test, were enrolled at three Royal Thai Army hospitals in Thailand. Key exclusion criteria were baseline AST or ALT concentrations of more than 3 times the upper limit of normal, cirrhosis, indication of paracetamol other than dengue infection, concurrent diagnosis of other causes of fever, or pregnancy. Patients were randomly assigned (1:1), by a computer-generated block randomisation procedure (block size of six), to receive either paracetamol (500 mg) or placebo (500 mg) every 4 h when body temperature exceeded 38°C during hospitalisation. Participants and investigators were masked to treatment assignment. The primary outcome was the proportion of participants with transaminase elevation, defined as serum aspartate transaminase (AST) and alanine transaminase (ALT) concentrations of more than 3 times the upper limit of normal on recovery day, in the intention-to-treat population. Prespecified interim analyses for safety and efficacy were performed with group sequential stopping boundaries. This trial is registered with ClinicalTrials.gov, number NCT02833584. Findings: Between Sept 1, 2016, and Dec 12, 2017, 125 participants were randomly assigned to receive either paracetamol (n=63) or placebo (n=62). 123 participants were included in the intention-to-treat population. The median daily dose of study medication was 1·5 g (IQR 0·8–2·0). The study was terminated early owing to a higher rate of transaminase elevation in the paracetamol group than in the placebo group (22% vs 10%; incidence rate ratio 3·77, 95% CI 1·36–10·46, p=0·011). The change of AST and ALT concentrations in the paracetamol group was higher than in the placebo group (mean difference 12·43 U/L per day, 7·16–17·71, p