End-of-life treatment preferences unknown in most ICU patients

Advance directives are rarely available, and shared decision-making between physicians and families in the patients’ best interest is the predominant model in deciding life-sustaining treatment (LST) limitations in Hong Kong’s intensive care units (ICU), a study has found.

Secondary analysis focusing on Hong Kong data from the prospective, global observational Ethicus-2 study on end-of-life practices in ICUs found that patients’ preferences regarding LST at the end of life are usually unknown. Only 0.4 percent of patients had advance directives, and only 15.8 percent made their LST wishes known. [Hong Kong Med J 2024;doi:10.12809/hkmj2310944]  

Of 4,922 patients admitted to eight ICUs in Hong Kong and screened between 1 September 2015 and 30 September 2016, 548 (11.1 percent) who had LST limitation (withholding or withdrawal) or died (due to failed cardiopulmonary resuscitation [CPR] or brain death) were included in the analysis. Follow-up continued until death or 2 months from initial decision to limit LST. Mutually-exclusive end-of-life categories included withholding LST, withdrawing LST, active shortening of the dying process, failed CPR, and brain death. If more than one limitation was triggered in a particular case, the most stringent limitation was chosen (ie, active shortening of the dying process was considered more stringent than LST withdrawal, followed by LST withholding).

LST limitation preceded deaths in 455 patients (83.0 percent), predominantly via withholding (353 patients, 77.6 percent) and less commonly via withdrawal (102 patients, 22.4 percent). Only 4 percent of patients with LST limitation were alive at 2 months. Of those who died without LST limitation, 80 (86.0 percent) had failed CPR and 13 (14.0 percent) were declared brain dead. No active shortening of the dying process (euthanasia) was reported.

Most patients (94.0 percent) lacked mental capacity for decision-making at the end of life. In the majority of cases (>86 percent), the ICU physician was involved in key aspects of end-of-life decision-making and implementation. The primary clinical reasons for limiting LST were unresponsiveness to maximal therapy (49.2 percent), multiorgan failure (17.1 percent) and neurologic dysfunction/failure (13.6 percent), while key considerations for decision-making were the patient’s best interest (81.5 percent) and perceived good medical practice (11.2 percent).

“High rates of LST limitation, such as those observed in this study, are aligned with international consensus, generally presumed to reflect good end-of-life practices, and have been associated with the presence of written end-of-life guidelines, such as those provided by the Hong Kong Hospital Authority,” the researchers noted.

“Compared with international practices, the time from ICU admission to LST limitation is relatively long in Hong Kong, possibly due to local cultural factors preventing earlier implementation of LST limitation,” they added.

Publication of results from this first large, multicentre study of end-of-life care practices in Hong Kong ICUs is timely, as the Hong Kong government is currently working on legislation related to advance medical directives and dying in place to enhance patients’ quality of life during their final days. [https://www.info.gov.hk/gia/general/202401/27/P2024012700723.htm?fontSize=1]