Scheduled lorazepam-based regimens superior to haloperidol alone for agitated delirium at end of life

Proactive use of scheduled lorazepam or lorazepam plus haloperidol is superior to haloperidol alone in managing refractory agitated delirium in patients with advanced cancer in the palliative care setting, according to results of a multicentre randomized clinical trial (RCT) conducted in Taiwan and the US.

In this double-blind, double-dummy, parallel-group RCT conducted between 16 July 2019 and 8 June 2023, a total of 111 patients with advanced cancer experiencing restlessness and/or agitation despite nonpharmacologic therapies and standard-dose haloperidol were recruited from three acute palliative care units in Taiwan and the US. These patients, stratified by site and Richmond Agitation-Sedation Scale (RASS) score, were randomized 1:1:1:1 to receive scheduled intravenous dose-escalated haloperidol, lorazepam, haloperidol plus lorazepam, or placebo every 4 hours until discharge, death, or withdrawal from the study. Follow-up duration was 30 days after medication administration. [JAMA Oncol 2025;doi:10.1001/jamaoncol.2025.2212]

The primary endpoint of change in RASS scores during the first 24 hours was assessed in 72 patients (mean age, 64 years; male, 58 percent). In all four treatment groups, RASS scores significantly decreased within the first hour and remained low during the next 24 hours.

At 24 hours, RASS scores were significantly lower in the lorazepam vs haloperidol group (mean difference, -2.1; 95 percent confidence interval [CI], -3.4 to -0.9; p<0.01) and in the combination vs haloperidol group (mean difference, -2.0; 95 percent CI, -3.2 to -0.8; p=0.002). No significant differences in RASS scores were observed between the haloperidol and placebo groups (mean difference, -0.5; 95 percent CI, -1.7 to 0.7; p=0.42) and between the combination and lorazepam groups (mean difference, 0.2; -1.1 to 1.4; p=0.79).

Patients in the lorazepam and combination groups required significantly less rescue medication for breakthrough restlessness or agitation than those in the haloperidol and placebo groups (37, 32, 56 and 83 percent, respectively; p=0.006).

No significant differences were found among the four groups in neuroleptic-related adverse events (AEs) between days 0 and 3, and in overall survival. The most commonly documented severe AEs were hypotension (26 percent) and hypoxia (26 percent), which were also changes characteristic of impending death.

No between-group differences were found in caregivers’ perception of the patients’ comfort or agitation. Bedside nurses, however, perceived a much higher degree of comfort (p=0.009) in patients who received lorazepam or the lorazepam/haloperidol combination.

“These results highlight three levels of sedation to personalize the management of agitated end-of-life [EoL] delirium, with haloperidol, lorazepam, or both to provide light, moderate, and heavy sedation, respectively,” the researchers concluded.

“Results of this trial raise important questions about the therapeutic target and strategy for treating EoL agitation,” wrote Dr Justin Sanders of the McGill University in Montreal, Canada, and Dr James Downar of the University of Ottawa, Canada. “While the aim in managing agitated delirium is to achieve a mid-range RASS score [ie, 0 to -2], patients in the lorazepam and combination groups both achieved a mean RASS score that was lower than the target, and significantly smaller proportions of them were in the RASS target range than patients in the haloperidol and placebo groups.” [JAMA Oncol 2025;doi:10.1001/jamaoncol.2025.2175]

“Although the assumption was that patients in the trial had terminal delirium, one cannot know if the more sedating interventions may have unnecessarily prohibited meaningful interaction among patients and caregivers,” they added.