Antimicrobial Stewardship in End-of-Life Care

Following the successful Irish Association for Palliative Care Seminar in February 2023, we publish here a summary of the winning presentation on the day. Cian Lannon and colleagues have written  a short summary for the BMJ SPC blog

Authors: Lannon C., Reilly L., Kennedy G., Fleming C., Waldron D.  University Hospital Galway, Saolta Group, Ireland  Email:



Antimicrobial stewardship is the promotion of the appropriate use of antimicrobials to improve patient outcomes, while also reducing the risk of patient harm and the emergence of antimicrobial resistance due to inappropriate use. Antimicrobial use in those approaching end of life (EoL) presents a more complex clinical decision given the difficulty with estimating prognosis, determining the proposed goal of antimicrobials (e.g. curative, symptom control) and accounting for additional potential harms. This audit in an acute tertiary hospital aimed to identify the prevalence of antimicrobial use in patients recognised as entering EoL and to review and improve rates of relevant documented discussions and decisions. Inpatients referred to palliative care and recognised as entering EoL were included for retrospective chart review, examining antimicrobial use during EoL, documented decisions and discussions. Between cycles there was an education campaign, through a hospital presentation and the regular consult service. The focus was on the importance of actively reviewing antimicrobial use on recognising EoL, including patients and their loved ones in discussions, and documenting these decisions and discussions in order to improve documentation rates. Initial results for cycle 1 showed in those receiving antimicrobials at recognition of EoL 42% were continued with 28% receiving antimicrobials until they died. Second cycle results showed 32% continued receiving antimicrobials during EoL but only 4% until death. Documented decision rates around antimicrobial use increased between cycles from 58% to 90% but with no documentation on goal of therapy.


There are many factors to consider with antimicrobial use at EOL. The goal of antimicrobial therapy should be clearly determined, as this can become more uncertain when approaching EOL. Antimicrobials can be used for symptom control, but this has varying rates of effectiveness for different infections. It’s important to remember that antimicrobial therapy at EOL is not without harm. It increases the risk of adverse drug reactions, drug-drug interactions and contributes to antimicrobial resistance. Intravenous antimicrobial therapy also involves the burden of intravenous cannulation and the associated intravenous fluid load. Involving patients and loved ones in discussions outlining these risks helps their understanding and gives an opportunity to explore their wishes and expectations. Results from this study show a significant antimicrobial burden of uncertain efficacy and potential harm, which may not reflect patient / family wishes. Following the education campaign there were earlier discussions and documented decisions, which can lead to better coordinated care between hospital staff, patients and their families.


Our results show a significant burden of antimicrobial use of uncertain benefit at a difficult time for patients and their families. International retrospective studies looking at antimicrobial use in EoL (not limited to those receiving palliative or EoL care) showed an even greater burden with 63.8% of patients receiving antimicrobials until the day of their death in a Korean study6, and an Australian study showing 62.7% receiving antimicrobials during EoL. Our patient cohort was receiving palliative care input explicitly for EoL care, which raises the question of the goal of antimicrobial therapy. The general goal of antimicrobial therapy is to cure infections, but even among high income countries infections remain a leading cause of death (lower respiratory infections alone represent the 6th leading cause of death) and so it may not always be possible or even appropriate to treat. Antimicrobial therapy may be used in an attempt to prolong survival even when disease is not curative, but in patients recognised as entering EoL the benefit of prolonging survival needs to be weighed against the risk of prolonging suffering9 and the increased risk of antimicrobial resistance. Also the effectiveness of prolonging survival at EoL is unclear, with some studies showing benefit but others showing no statistically significant difference in survival time when antibiotics are used or withheld in accordance with patient wishes at EoL.

Antimicrobial therapy can also be used primarily to relieve symptoms, but this is often ineffective. A systematic review of antimicrobial use for symptom management in palliative care patients found considerable variation with the best symptom response seen in urinary tract infections (60-92%) compared to respiratory tract infections (0-53%), and no symptom response in treated bacteraemia.

Antimicrobial use in EoL presents a complex decision incorporating difficulties estimating prognosis, balancing uncertain benefit against additional harms, public health concerns and patient / family wishes. Frameworks exist to help with this decision-making, including algorithms looking at evaluation of the goals of care followed by assessment of whether symptomatic benefit is likely if chosen9. Our review showed that this process is not always carried out and documented in acute hospitals, but a greater focus on this issue leads to more discussions and documented decisions. Conversations with patients and their loved ones about potentially withdrawing antimicrobials can be challenging but ultimately inform and empower them to be part of their care. Documenting these discussions and decisions is crucial to coordinate care between extended hospital teams. Ultimately achieving the best outcome requires an individualised approach, and active, inclusive decision making.

(Visited 379 times, 1 visits today)