Background Palliative sedation therapy (PST) is used to treat refractory symptoms at the end of life. Guidelines surrounding commonly used medications for PST are available, but variability exists for subsequent agents including phenobarbital. Doses of phenobarbital loading doses and continuous subcutaneous infusion (CSCI) doses are varied, as is the recommended route of administration and use of concurrent PST.

Objectives Identify the indication, dosing range and time to death from phenobarbital CSCI commencement

Identify the use of preceding, and concurrent PST medication choices

Methods This retrospective clinical audit investigated inpatient phenobarbital CSCI use in the palliative care setting from 1 January 2021 to 22 April 2024. Data was extracted from the electronic medical records system, and collected in a spreadsheet detailing patient characteristics, diagnoses, dosing, timing, and indication for midazolam, levomepromazine, phenobarbital and propofol.

Results 86 patients were identified initially. Five were excluded as phenobarbital was a pre-existing prescription for seizure prevention, thirty-nine were prescribed but not administered phenobarbital, fourteen prescribed and administered phenobarbital on an as-needed basis only, and five were administered in a CSCI for seizure control not PST. 23 patients were included following exclusion.

The most common indication listed for phenobarbital use was ‘agitation’ (n=21, 75%), followed by ‘seizures’ (n=5, 18%) and ‘sedation for NIV withdrawal’ (n=2, 7%). More patients had a malignant diagnosis listed as their cause of death (n=14, 58%) than a non-malignant diagnosis. All patients had the phenobarbital CSCI running at time of death, and were described as comfortable at death. No adverse reactions were documented.

The initiation dose of phenobarbital ranged from 0–200mg subcutaneously (SC), with the most common dose being 200 mg (n=17, 74%). The commencement CSCI dose ranged from 400–1200mg per 24 hours, with the most common starting dose being 800mg (n=8, 35%). The maximum dose used was 1800mg per 24 hours. The average time to death from phenobarbital CSCI commencement was 45 hours. Higher established doses of phenobarbital did not appear to result in a shorter time to death.

Prior to phenobarbital CSCI commencement, all patients had a midazolam CSCI running (mean dose 50mg, range 10–80mg), which was continued in fourteen (60%) patients. Seventeen (74%) patients had a levomepromazine CSCI running (average dose 137mg, range 25–250mg), which was continued in twelve (76%) patients. After phenobarbital commencement, eight (35%) patients had both a midazolam and levomepromazine CSCI running.

Conclusion Within our centre, phenobarbital was infrequently used and most commonly prescribed for agitation. It appears to be an effective medication for refractory symptoms at the end of life. There was a wide range of phenobarbital dosing, and inconsistent prescription of concurrent PST agents. Future studies could expand on appropriate phenobarbital prescribing by considering multi-centre analysis and mixed method studies on the role of other PST agents in conjunction with phenobarbital.